Brightime Intrnational Ltd. & Another v. Lap Fung Fire Engineering Co. Ltd

DCCJ5977/2002

IN THE DISTRICT COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

CIVIL ACTION NO. 5977 OF 2002

__________

BETWEEN

__________

Before H.H. Judge Muttrie in Court

Date of Trial :  1 – 3 & 6 September 2004

Date of Judgment : 22 September 2004

JUDGMENT

1.The plaintiffs claim damages from the defendant for negligence in the installation of an automatic fire sprinkler system in their premises. A sprinkler head burst after installation causing a flood and consequent damage to the plaintiffs’ goods. The claim is for $471,227.63 for loss of goods plus $23,280.00 being loss adjuster’s fees. In fact the plaintiffs have been compensated by their insurer and they are now seeking recovery on the insurer’s behalf.

2.The plaintiffs are related companies which trade in cosmetics including fragrances and aromatherapy products. At the relevant time the 1^st plaintiff was the tenant of premises at Unit A2, Kaiser Estate Phase 1, Hung Hong which the two plaintiffs used as a warehouse with a small office. The 1^st plaintiff’s landlord itself occupied the other part or parts of Unit A which are described as A and A1, or simply as Unit A.

3.Kaiser Estate is equipped with a fire prevention sprinkler system. The defendant is a registered fire service installation contractor, and since 1999 it had been appointed by the management company of the Kaiser Estate to conduct annual checking of the fire prevention system there. The defendant was therefore also commonly engaged by tenants of the estate for fire prevention system maintenance work in their premises.

4.Following receipt of a notice from the Fire Services Department requiring the plaintiffs’ landlord to rectify a fire hazard in Unit A by removing a false ceiling which obstructed the sprinkler heads or alternatively to extend them below the false ceiling, the landlord appointed the defendant to carry out this work in its own premises and those of the plaintiffs.

5.It is not in dispute that in the landlord’s premises, 39 sprinkler heads had to be installed, including 14 units additional to those already existing; and in the plaintiff’s premises 38 sprinkler heads had to be installed, including 11 additional units. Pipework for the new heads, as well as for the relocation of the existing heads had also to be installed.

6.The work was done some time between 9 and 12 October 2000; when it was completed is a matter in dispute. On the evening of 12 October, one of the sprinkler heads in the plaintiff’s premises burst, causing loss and damage to the goods which the plaintiffs kept there. The burst occurred some time after the plaintiff’s staff had left the premises and the flooding was discovered at about 9 p.m. the same night.

7.There is no contractual relationship between the parties. The plaintiffs claim is in tort. The plaintiffs plead that the defendant was negligent. The particulars of negligence are set out at length but in short the main points are that there was no or no proper testing of the system after its installation which was completed on 12 October 2000,  and the burst was caused by defective material of the sprinkler head or poor workmanship on the part of the defendant. The plaintiff also relies on res ipsa loquitur. The defendant denies negligence. It says that the system was properly installed and tested on 10 October and that the heads were of good and sound condition. There was flammable material in the premises which could have caused the sprinkler bulb head to burst by reason of heat. There are averments of contributory negligence against the plaintiffs and the landlord although these were not pursued at trial and in any event the landlord is not a party to these proceedings.

8.The sprinkler system with which we are concerned here was divided into five zones within the building. The zone concerned here included the whole of the second floor. In each zone, water was supplied to the sprinklers from a tank on the roof. The water was under pressure; the lay witnesses say the pressure was 120 pounds or 120 kilograms and the experts say is was 120 kilopascals, but the figures are not important here; the point is that it was under pressure, and held in the pipes by the sprinkler heads. For work to be done on the system in any zone the water had to be released by opening a valve situated on the ground floor of the building. Once the work was completed the system would be charged up again by closing that valve, opening another to let water in and turning on the pump to bring the water up to the required pressure. This last was, according to one of the defence witnesses an automatic process. The water went in at a pressure of 70 pounds (or whatever unit it was) and the pump brought the pressure up to about 130; it would then settle back to 120 after the air in the system came out of it. When “testing” is referred to, it seems that this is what is meant.

9.The sprinkler heads concerned here are much the same as those which can be seen protruding through the ceilings of my courtroom and those of many other buildings in Hong Kong. The heads were manufactured by Spraysafe Automatic Sprinklers Ltd. in the United Kingdom and supplied to the defendant by Henry Engineering and Supplies, which, according to its assistant sales manager has 70% or 80% of the market for sprinkler heads in Hong Kong.

10.The type of head is approved by the Fire Services Department. The head itself consists of a short metal threaded tube with two arms coming out of it to an end piece with a rosette screwed into it. The tube is screwed vertically on to the end of a water pipe. Within the downstream end of it is a stopper, held in place by a little cylindrical glass bulb which lies between the two arms and is pressed up against the stopper by the screw of the rosette. The glass bulb contains a liquid. When the bulb is subjected to a temperature of 68°C, the liquid expands and breaks the glass; this in turn releases the stopper and allows the water to be discharged under pressure against the rosette which disperses it over a wide area.

11.The sprinkler head concerned here has been produced as Exhibit D1. The stopper and glass bulb are gone but the metalwork is intact. The defendant’s director, Mr Chen, produced the head in the course of giving evidence along with a new, intact head produced as D2. The exhibits had not been the subject of discovery, but no point was taken on that nor was there any dispute that the exhibit is the sprinkler which burst. What happened here, as may be seen from the photographs also produced, is that the glass bulb broke, just as it would have done if it had been exposed to heat. But the cause of the breakage is not known. No one was there to see it. We are left mainly with opinion evidence as to what the cause might have been. That opinion evidence comes from two experts who were consulted long after the incident. They did not have the advantage of a being able to carry out a prompt investigation. The investigation was done by the loss adjuster, Mr Ip and the experts had to rely on his evidence of what was found at the scene.

12.In brief the case comes down to this. The experts generally agree that the possible causes of a burst sprinkler bulb are fire, defective material, poor workmanship or external impact after installation. Both agree that defective material and external impact are unlikely. In effect the plaintiff’s expert rules out fire and says that poor workmanship is the most likely cause. The defendant’s expert rules out poor workmanship and says that fire is the most likely cause. The factual issue of whether the work was completed and tested on 10 October, or only completed on 12 October is relevant to the question of causation. I will deal first with this question.

13.Evidence for the plaintiff comes from its director Ms Iris Ho, its insurer’s loss adjuster, Mr Ip, and an employee, Ms Ng. In brief Ms Ho did not work at the premises; she worked in an office in Tsim Sha Tsui and she was told by her staff in the premises, in particular a supervisor called Danny, of what was happening there. She says that the landlord wrote to her to the effect that the work was to be done between 9 and 11 October. In fact it lasted for about three days. On 12 October Danny telephoned her and told her that the work was finished. He did not tell her when or if the system was tested. She remembers the date because it was the same day on which she was called in the evening by the management office of the building, and told that there was a flood in the premises.

14.In fact the plaintiff’s first letter to the landlord about the incident, dated 17 October 2000 indicated that the sprinkler had been installed on 10 October. Ms Ho was asked about this and her response was that she had only mentioned the installation and not the period it took.

15.Various subsequent letters emanating from the plaintiff’s solicitors (different firms at different times) and loss adjusters acting for the management’s and the plaintiffs’ insurers also indicated that the work had been done on 10 October. Indeed that was the plaintiffs’ stance from the outset, until the Statement of Claim was amended in June 2003 to plead that the work was finished on 12 October. Most of these letters came from persons whom the plaintiffs were not instructing directly (later solicitors being retained by the insurer) but the first solicitor’s letter came from a firm directly retained. Ms Ho admitted that the contents were correct and admitted having instructed the solicitors to write the letter.

16.The other witness, Ms Ng was a general worker who cleaned Ms Ho’s office in the mornings and went to work in the premises in the afternoons. She said that the work took about three days, but her first statement was very recent and provided only from memory. She was not sure what the workers were doing; at most she could remember that they had used step ladders. She did not know who the workers were. She obviously had little recollection of the matter.

17.The loss adjuster, Mr Ip, who was called for the plaintiffs, had no knowledge of when the work had been done but says that when he went to the scene on the morning of 13 October a representative of the defendant told him that the work in Unit A2 had been finished on 10 October, but that the defendant was still working in the other part, i.e. Unit A. He asked Ms Ho when it had been finished, and she said that the installation lasted one or two days and at the time when the accident happened the installation should have been completed.

18.The plaintiff called on subpoena a Mr Lam, the responsible officer of the management company of the Kaiser Estate, to produce documents. Mr Lam was not working at Kaiser Estate at the relevant time. The first of these documents is a letter from the defendant to the management company dated 7 October 2000 advising it that the modification works in Units A and A2 as well as commissioning and testing of the sprinkler system would be carried out on 9 October to 13 October 2000. Such a letter is said to be always necessary when the defendant has to carry out this kind of work. The next was a set of security officer’s patrol records which show that the defendant “released water and said that 2/F A is changing sprinkler” on the morning of 10 October, with two further entries for release of water on the mornings of 11 and 12 October. This apparently refers to the draining of water out of the sprinkler system for the zone covering the 2^nd floor. Documents were also produced which show that the defendant notified the management that on 22 September 2000 it would modify, commission and test the hose reel system, and on 17 October 2000 would modify, commission and test the sprinkler system in Unit 2C. Job orders and quotations for these jobs were also put before the court.

19.Evidence for the defendant as to when the work was done came from first from Mr Chung, its former marketing manager. He was the author of the notification letters referred to above. He said that it was the defendant’s practice to state more days of work than would actually be required. In fact the work had been finished on 10 October. Under cross-examination he said that it was the practice, in respect of all or most of the units, to charge for two water releases, whether one or two were needed. He had quoted three releases for Unit 2C, and seemed to be saying that work in that unit was carried on for more than the two days notified to the management; but ultimately he said that water had been released in respect of that unit only on 22 September and 17 October; any water release required a notification letter to the building management. 

20.The defendant’s director, Mr Chen also said that the work in the plaintiffs’ premises had finished on 10 October. He had told the loss adjuster, Mr Ip, this when he was asked about it on 13 October. He gave evidence of how he estimated and charged for a job, and of Mr Chung’s estimate of how he arrived at the five days needed for the work. He explained how his subcontractor, Mr Lam, and his two workers did the work and under cross-examination gave lengthy explanations of what exactly had to be done. So far as the water releases on 11 and 12 October were concerned his evidence was in effect that his workers took advantage of the notification that work would be done in Units A and A2 up to 13 October, and hence that the water could be released on those days, to release the water and let them do some work in Unit 2C. He also said that he had been there when the work was finished and the system tested. Staff of the plaintiffs were present, but not the responsible person.

21.The defendant’s sub-contractor, Mr Lam is the person who actually did the work, along with, according to him, his two workers. He says that he did all the preparation work on 9 October and all the installation on 10 October, finishing by about 5 p.m. He then tested the system and checked that there were no leaks in Unit A2 in the presence of the staff there. He finished by about 6 p.m. From cross-examination, however, it appears that he did all the preparation for Unit A on 9 October and only started installation there after he had collected the new sprinklers from the supplier on the morning of 10 October. (The witness from the supplier says that the sprinklers were collected about noon that day). Thereafter he did all the preparation and installation for Unit A2 in the afternoon of 10 October. He tested from the control room; he went back and forth from there to A2 to check for leaks. On the following day he got the responsible person of the landlord, a Mr Ho Lim Keung, to sign and chop the job lists to accept that the jobs were finished in both units. But the job lists, although signed are not dated.

22.The evidence for the plaintiff on this matter comes from Ms Ho, who got all her information from staff at the premises, and not from her own knowledge, and Ms Ng who is operating entirely from hazy recollection of what happened four years ago. Against this, we have the position originally taken by the plaintiffs and those advising or representing them that the work was finished on 10 October. I do not think it can be said that the plaintiffs only amended because of what their expert said; the amendment pre-dated the expert’s report. The later communications from solicitors and insurers cannot necessarily be taken as indicating that Ms Ho is not telling the truth. However it is difficult to get round the fact that in on 17 October the plaintiffs’ original solicitors said that the work was finally completed on 10 October, on instructions received from her. It is also difficult to see that if Mr Ip asked her, as he says he did, when the work was finished, he having already been told by the defendant’s representative (presumably Mr Chen) that it was finished on 10 October, while she knew that it had been finished only on the previous day, she would not have told him so.

23.On the other hand we have the evidence given for the defendant which is also unsatisfactory. On any sensible reckoning, Mr Lam cannot have taken a day and a half to prepare and install the sprinklers in Unit A, and only an afternoon to prepare, install and test them in Unit A2, given that on his own admission the scope of the work in both units was not much different. This must be so even taking into account that three men were doing the job. A fortiori when one considers that for 11 of the sprinklers, new piping had to be installed, as well as extension piping for the other 25. Mr Lam’s estimate is that it took “a few minutes” or “ten minutes including the piping” for an easy job e.g. relocating the old sprinklers and 10-odd to 20 minutes for the more difficult ones. Nothing has been said about removing, replacing and/or drilling out ceiling boards but if at least part of the job involved bringing the sprinklers down through the boards, that would have come into the equation as well.

24.There is also the opinion of the plaintiffs’ expert, Professor Chao that the whole job would take more than a day although this has to be viewed against the fact that he is now an academic and has not worked with sprinklers in the field for many years. However, even without this opinion, on the basis of what Mr Lam himself says, it seems most unlikely that even with three men working the of the preparation and installation in Unit A2 could be finished in an afternoon.

25.Further, if the whole job is both units was only going to take two days, there is no clear explanation as to why the notification to the management, which was done after Messrs Chung, Chen and Lam had all visited the site and seen at least some of what the job required (without looking behind the ceiling boards) should have said that it would take five days. It is all very well to say that they wanted to give themselves leeway but a fire prevention system should not be out of commission for longer than necessary. Then we have the water release on 11 and 12 October. This is not in dispute; Mr Chen says that it was to for work in Unit 2C but this does not accord with the notifications to the management about the work to be done in that unit. It is possible of course that the defendant took advantage of the notification in respect of Units 2 and 2A to turn off the water and do some work in Unit 2C; the story is not implausible; but one might have expected some job sheets or other supporting evidence. Also, it appears that Mr Ip, the adjuster, was not specifically told about 2C; he was told that work was still going on in the landlord’s unit.

26.I accept however that the defendant, unlike the plaintiff has always been consistent, even in the initial report to Mr Ip that the work was finished on 10 October. Also Messrs Chen and Lam are speaking from their own knowledge which Ms Ho is not. There is also the fact that someone from the landlord signed the job sheets. Mr Lam says that this was on 11 October. Though the documents are not dated it hardly seems likely that anyone would have signed them after the accident had happened, and the question had therefore arisen of who, or whose insurer, was going to pay for it.

27.I turn to the question of causation. The plaintiff’s expert, Professor Chao, puts forward four possible causes for a burst of this sort. The first is fire and the second is a heat source such as a radiative heater. Either of these must bring the temperature of the air surrounding the bulb to 68°C before it will burst. The third possibility is that the sprinkler is hit and damaged at the bulb and the fourth is that the water leaks from the bulb due to poor workmanship or defective material. However this last will usually be detected during the testing phase, or very soon after the establishment of the “all wet” condition of the system; i.e. very soon after the pressure is turned on.

28.Professor Chao rules out fire or heat as a possible cause. Likewise, because there is no evidence of impact damage, and the sprinkler was situated above an office desk, in an area where it was unlikely to be hit by anything, he rules out that possible cause. On the basis that he does not believe that the job could have been finished on 10 October, Professor Chao concludes that it must have finished on 12 October and that the leakage started shortly after the pressure was turned on on that day and he gives his opinion that it was poor workmanship of or defective material supplied by the defendant which caused the burst.

29.The defendant’s expert, Mr Ng agrees generally as to the possible causes but rules out  defective material or poor workmanship, pretty well on the basis that in the event of either, the head could be expected to burst once the system was pressurised. However says that fire is a possibility even if no residue was found.

30.It seems to me that external impact can be ruled out because there is no direct evidence of it and also because although there were places in the plaintiffs’ premises where articles were put on shelves near the ceiling, the sprinkler which burst was over a desk and there was nothing close which might have impacted. Of course a workman might have it the sprinkler after installation with a ladder or the like but if that happened the break would most likely be immediate.

31.The experts agree that defects in manufacture are very rare. In any event if there were any such defect it would probably appear as soon as the system was pressurised.

32.This leaves fire and poor workmanship on the defendant’s part. I think it is necessary to consider fire first.

33.The experts agree that for a fire to start, there must be fuel, oxygen and something to provide ignition. The plaintiff’s expert, Professor Chao, rules it out. According to the plaintiff’s loss adjuster, Mr Ip, and the documents provided to him by the plaintiff and the landlord do not show any evidence that a burning occurred. In order to burst the sprinkler, the air around it would have to be heated to 68°C. Air at normal temperature would mix with the smoke and flame rising from any fire, and it would take at least a couple of minutes for the resulting mixture to reach that temperature. A fire burning for that length of time would inevitably leave traces in the form of charred or burnt material, or soot on surfaces.

34.Mr Ng however raises the possibility of a fire caused by ignition of vapour. There were inflammable materials inside the premises. The plaintiffs kept aromatherapy oils in the premises as well as cosmetic products containing alcohols. Vapour given off by these oils and other products could mix with oxygen in the right proportion for burning. The only missing factor was a heat source, which could be a cigarette end left by the occupants before leaving the room or any hot surface heated by electricity. Energy as small as a spark could produce a fire. Mr Ng accepts that once a fire started it would need to burn for a couple of minutes to heat up the air sufficiently to break the bulb. However he says that the resulting deluge could remove the traces of any soot or other fire residue.

35.It seems to me that fire is unlikely. No doubt it is theoretically possible to set off the alcohol or other volatile vapour that might come off cosmetic products or aromatherapy oils, if the right mixture of fuel and oxygen could be obtained. Professor Chao says, on the basis of what he has been told by a chemist colleague, that it is possible for the oil to burn if one puts a flame right on the surface; but if the igniter were 10 centimetres away nothing would happen. But aromatherapy oils may be and often are heated over a flame and one does not hear of them catching fire like the brandy on a pan of crêpes Suzette, as it is heated over a flame when the waiter prepares the dish by the customer’s table in a fancy restaurant. Nor does one hear of conflagrations in cosmetic stores, or explosions in them of the sort which may blow up a flat where there is a gas leak and someone presses the doorbell and a spark sets it off. Nor indeed is there any expert evidence of such fires or explosions having occurred in the past. I prefer the evidence of Professor Chao that if there had been a fire, some residue of it would have been found. So I think that fire can be ruled out.

36.This leaves bad workmanship. Here again we are largely in the realm of speculation. The suggestion seems to be that the head could have been dropped, or stressed while being screwed on to the pipe with excessive force, or that the bulb could have been displaced or the arms of the sprinkler could have been distorted by the fitter so that the bulb was displaced. I have been referred to the judgment of Milmo J in Camis v Matthew Hall Mechanical Services Ltd. [1969] 1 Lloyd’s L.R. 43, another case where a sprinkler head burst without any known cause, where these possibilities are referred to at page 45. Unfortunately I did not hear expert evidence on “point stress” and other matters referred to in that judgment and I cannot rely on it for evidence of fact or opinion.

37.What I do have is the evidence of the defendant’s witness Mr Cheung, a retired Chief Technical Officer of the Fire Services Department, who says that if there was any damage caused during fitting – and he appears to accept that such damage could be caused, without actually breaking the bulb – the bulb would immediately burst on testing, when water pressure was applied to it.

38.Professor Chao also said that poor workmanship would usually be found on test. Under cross-examination, he said that the chance of damage in installation causing a burst after two days would be very low. He could not think of any other reason for a burst on 12 October.It seems that it is largely because of this that he did not believe that the job could have been finished on 10 October.

39.It is argued for the plaintiff that I should reject the defendant’s evidence that the job was finished and tested on the afternoon of 10 October. It was the plaintiff’s evidence (from Ms Ng) that the staff left the office on 12 October at 5.30 p.m. It was likely, because Mr Lam said that he worked until 6, that the pressure was turned on in the absence of the plaintiffs’ staff and that the head then burst.

40.There are difficulties on both sides concerning the question of when the job was finished. I have outlined them above. The plaintiffs cannot explain away the fact that they took the position in the beginning that the job was finished on 10 October. But the defendant cannot explain away the water releases on 11 and 12 October. The story of work in Unit 2C is unlikely and not supported.

41.There was no fire, and no chance of post-installation impact by anyone other than the defendant’s men so there must have been something wrong with the sprinkler head or it would not have burst. A manufacturing defect is regarded as highly unlikely and in any event is not pleaded, nor does the defendant plead a latent defect, i.e. one which could not be detected by the use of reasonable care and skill. The evidence on both sides is that if there is some damage to a sprinkler head, it is much more likely to burst when the pressure is put on, than some time afterwards. If the job had been finished on 10 October the head would have been subjected to pressure on that day when Mr Lam turned the pressure on, and again on 11 October when the pressure was turned on, as it must have been if the system had to be drained again on 12 October. So it seems more probable than not that the work was not finished, and the burst sprinkler head not subjected to pressure until some time on 12 October.

42.What was done about testing or inspection on that day is not clear. The only test which the defendant could carry out is that of turning on the pressure and inspecting to see there are no leaks. Ms Ho says that Danny the supervisor told her, definitely within business hours, that the work was finished. He did not say, and she did not ask him, whether there had been any test; but the fact of finishing itself implies turning on the water at pressure; and we know that was done because the flood ensued. It follows that the work must have been finished and the system turned on some time before 5.30 or 6 p.m. and that Danny knew about it. The sprinkler cannot have burst within business hours or Danny or some other staff member would have known about that too. So it cannot be the case that there was no proper test, or no test carried out in the presence of the plaintiffs’ staff.

43.What we have, then, is a burst occurring within a fairly short time, say a couple of hours, after the system was pressurised; and the cause of the burst is unknown. The plaintiffs seek to make the defendants liable by eliminating all other causes save negligence on the part of the defendant, or alternatively to rely on the maxim res ipsa loquitur.

44.For the maxim to apply the cause of the accident must be unknown and one which would not normally happen without negligence on the part of the defendant who is in control of the object or activity which injured the plaintiff or damaged his property. Once it applies, the onus does not shift to the defendant to disprove negligence but, in the words of Milmo J in Camis v Matthew Hall, above, at page 47,

“….the defendants must…in order to escape liability, prove that the bursting was due to a specific cause which did not connote negligence on their part but which points to its absence as more probable; or, alternatively, if they were unable to do this, they had to prove that they used all reasonable care in and about the manufacture and fitting of this particular spreader, together with the ampoule which it held.”

45.The defendant argues that the maxim should not apply because the plaintiffs have called expert evidence to eliminate all probable causes and conclude that the cause is negligence. So the cause is not unknown to the plaintiffs. Further there are several possible causes, only one of which involves negligence, and the defendant was only in control of the installation and not the premises themselves.

46.It seems to me that, having got to the position where fire, manufacturing defect and post-installation impact can all be ruled out, the maxim must apply. The defendant was in control of the installation. Shortly after installation, the head burst. Once the other probable causes are eliminated the only one remaining is negligent handling by the defendant or its agent or servant, in the installation. The defendant did not, as in the case cited, manufacture the head but must prove that it used all reasonable care in and about the installation of it.

47.In my view the defendant has not done so. There is no evidence from the person who actually did the fitting; Mr Lam cannot say who that was, between himself and his two workers. Mr Lam can say that he brought the heads, in their packaging from the shop where he bought them, but he cannot thereafter trace what happened to the individual head before it was finally screwed on to the pipe; so there is no positive evidence that it was not dropped, or subjected to some other impact in the fitting process. I take the point that there is no visible distortion of the arms; this may rule out a prolonged or heavy twisting of the unit by the arms if the fitter’s spanner slips (for the unit is made with two flats, and may be turned with an open-ended spanner or even an adjustable wrench; mention has been made of a special tool but there is no evidence of what that might have been) but it does not rule out an impact caused by such slippage.

48.The point is that what the defendant would have to do in order to escape liability is to produce the fitter to say that he took the head out of its packaging, and used all reasonable care in fitting it; but this the defendant cannot do. So it follows that I must infer negligence on the part of the defendant and find for the plaintiffs on liability.   

49.I turn to the question of quantum. The only real dispute on this relates to the salvage value, which the defendant says should be deducted. The adjuster suggested to the plaintiff that some of the claimed items could be reconditioned and repacked for sale, and finally agreed a salvage figure of $50,845.66. But Ms Ho says that in any event all the goods were unsaleable and the figure was only agreed in the interest of speedy resolution. I can accept that the goods are such that they have to look good on the shelf, or they will not be bought; so even a defect in packaging might render an item unsaleable. But there was evidence from Ms Ho that the plaintiffs did in fact repackage some of the lotions and the like into smaller bottles. It is difficult to see why the plaintiffs would have accepted the salvage figure if no salvage, even by repackaging in this way, had been possible. So I accept that the salvage figure must be deducted.

50.There will accordingly be judgment in favour of the plaintiff for $471,227.63 - $50,845.66, i.e. $420,381.97, plus the adjuster’s fee of $23,280.00 which produces a grand total of $443,661.97, plus interest thereon, as suggested by counsel at half the judgment rate from 12 October 2004 until judgment and thereafter at the judgment rate, and costs to be taxed if not agreed. Counsel have agreed that the costs order should be final even though the judgment is to be handed down.

Mr. Jonathan Wong instructed by Messrs. Tsui & Co. for the Plaintiff.

Mr. Lawrence Yip instructed by Messrs. David Ho, Kevin Kong & Co. for the Defendant.