[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales High Court (Commercial Court) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Commercial Court) Decisions >> Glencore International A.G. v Metro Trading International Inc & Ors [2001] EWHC 490 (Comm) (01 August 2001) URL: http://www.bailii.org/ew/cases/EWHC/Comm/2001/490.html Cite as: [2001] EWHC 490 (Comm)

[New search] [Printable RTF version] [Help]

IN THE HIGH COURT OF JUSTICE
QUEENS BENCH DIVISION
COMMERCIAL COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		1st August 2001

Mr Justice Moore-Bick:

Introduction

1. In Phase 1 of this litigation I determined a number of questions of principle relating to the law applicable to the passing of title to oil held in storage in Fujairah. In summary, I held that under English conflicts of laws rules the transfer of title to movable property is governed by the law of the place where the property is situated, that being in the present case the law of Fujairah. My judgment is now reported at [2001] 1 Lloyd's Rep. 283.
2. Phase 2 is concerned with the contractual arrangements between Metro Trading International Inc ("MTI") and the companies for which it held oil in its floating storage facility at Fujairah from time to time, principally Glencore International A.G. ("Glencore"), and the effect under the law of Fujairah on title to oil held in storage of various dealings with that oil in the context of those contractual arrangements. In addition to Glencore three other companies, Caltex Trading Pte Ltd ("Caltex"), Mobil Export Corporation ("Mobil") and Arexco International Ltd ("Arexco") entered into agreements at one time or another with MTI under which they delivered oil products to MTI for storage. A fourth company, Texaco International Trader Inc. ("Texaco"), entered into an agreement with Metro Oil Corporation ("MOC"), an associated company of MTI which operated a refinery and oil storage facilities in Fujairah, for the processing of crude oil. Products delivered to Texaco under the agreement were usually supplied from the storage vessels operated by MTI and Texaco maintains that MOC used MTI's facilities to store product belonging to Texaco. The relationship between MTI and each of these companies, which have been referred to in this litigation as "the oil claimants", differed in certain respects, but they have all laid claim to the oil which remained in the floating storage at the time of MTI's collapse in early 1998. In each case it will be necessary to consider the precise terms of the contract between MTI and the relevant oil claimant as well as the course of dealing between them to ascertain to what extent, if at all, they are entitled to assert rights of ownership against the remaining oil. It was apparent from an early stage that in the case of Glencore this would involve a complex and time-consuming enquiry. At the pre-trial review, therefore, I directed that at the first trial under Phase 2 the court would only deal with issues relating to the relationship between Glencore and MTI and that the relationship between MTI and Caltex, Mobil, Arexco and Texaco would be the subject of a second trial currently fixed for hearing later this year.
3. As well as making a claim against MTI asserting a proprietary interest over the oil which remained in its possession in February 1998 and damages for the wrongful disposal of oil no longer held in storage, Glencore has made a claim against its insurers in respect of the loss of the oil which it alleges MTI wrongfully disposed of. This claim is resisted on the grounds, among others, that much of the oil in respect of which Glencore seeks to recover was disposed of with its consent and that to that extent there was no loss under the policy. Any consideration of the relationship between Glencore and MTI therefore inevitably raise issues of fact and law which are relevant to the claim that Glencore is making against its insurers. The insurers were therefore given permission to be represented at the trial and took a full part in the proceedings. At the beginning of the trial their position was one of cautious neutrality: they had put in issue the cases advanced by both Glencore and MTI, but had specifically stated in their pleadings that they did not seek to advance any positive case of their own. During the course of the trial, however, it became increasingly apparent that they were adopting a positive stance in support of a case which was materially different from that of either MTI or Glencore. This case, which became known as 'the third way', was eventually reduced to a formal pleading and in the absence of any serious objection by other parties I gave the insurers permission to make the necessary amendment.
4. Two other groups of litigants were also represented at the trial. The first ("the shipowner defendants") comprised the owners of the vessels Shoko and Cherry on which MTI had shipped cargoes of oil in which Glencore claimed to hold title. In each case Glencore had brought proceedings claiming damages for wrongful interference with its goods. The other ("the Fal defendants") comprised two associated companies, Fal Oil Co. Ltd and Fal Shipping Co., against whom Glencore made similar claims. Each of these groups of defendants relied in their defence on the acquisition by MTI of a good title to oil delivered to them, or, in the alternative, on MTI's having authority to dispose of the oil in question. To that extent their cases raised issues which were the same as, or very closely related to, those raised by Glencore's claim against MTI. In these circumstances, although each of these groups put forward arguments of their own, to which I shall come in due course, they both adopted the case advanced by MTI and as a result played a rather less significant role in the trial.

The nature of the dispute

5. The questions identified for determination within Phase 2 are set out in an appendix to this judgment. It is unnecessary to repeat them here, but it may be of assistance to describe the broad outlines of the dispute between the parties before embarking on a detailed consideration of its various aspects. From about 1988 MTI (then called Metro Bunkering and Trading Company) carried on operations in Fujairah as a supplier of marine bunkers. The company was part of the Mavrakakis group and at the outset its operations were confined to the supply of bunkers to the Mayamar fleet operating in the area of the Gulf. However, it gradually extended its operations to the commercial bunker market. Cargoes of high viscosity oil, gasoil and cutterstock were purchased in the locality and used to blend fuel oil suitable for use as bunkers. This involved the establishment of a floating storage and blending facility using vessels anchored in the waters off Fujairah. By March 1993 the business had expanded to the point at which MTI was able to employ as its primary storage and blending vehicle the Metrotank, a vessel of some 300,000 dwt. capacity which was specially adapted by MTI for that purpose. In addition other vessels were employed to increase the storage and blending capacity as circumstances required.
6. Glencore and MTI were introduced to each other through an oil broker in March 1992 when Mr. Arie Silverberg, then head of Glencore's oil department, and Mr. Jean-Claude Heuzé, one of Glencore's fuel oil traders, met Mr. Ed Griffin of Metro Trading (USA) Inc., the company which acted as MTI's representative in the United States. Between April 1992 and February 1998 Glencore and MTI did business together on very many occasions and as their relationship developed so did the nature and extent of the business. In the event the vast majority of the transactions between them were carried out under joint venture agreements which have since become known for convenience as 'JV1' and 'JV2'. The precise nature of their relationship changed significantly in June 1996 when the parties moved from JV1 to JV2, but from about the end of 1994 one constant feature of the relationship was the delivery by Glencore into MTI's floating storage facility at Fujairah of substantial quantities of oil for subsequent sale, often in the form of blended products, into both the local bunker market and the fuel oil market in the Indian sub-continent and the Far East, especially Singapore. The first stage of this relationship between Glencore and MTI, JV1, was in the nature of a true joint venture under which Glencore and MTI shared the profits and losses arising from the sale of oil brought into the venture. The second stage, JV2, although treated by the parties as a continuation of JV1 and still described by them as a joint venture, differed in that Glencore received a fixed return on all oil brought into the venture and MTI assumed the entirety of the market risk. Under JV2, therefore, MTI retained the whole of any net trading profit and bore the whole of any net trading loss.
7. Oil required for the operation of the joint venture was purchased from prime suppliers by both MTI and Glencore, but it was an essential element in the arrangement that Glencore would finance the supply of oil. Accordingly, whenever MTI entered into a contract with a prime supplier Glencore entered into a contract with MTI to purchase the oil on back to back terms and Glencore provided the letter of credit necessary to enable the prime supplier to receive payment. Initially it was the practice for Glencore to open its own letter of credit in favour of MTI which in turn opened its own letter of credit in favour of the prime supplier, but later it became the practice for Glencore to open a letter of credit in the name of MTI in favour of the prime supplier and for MTI to give irrevocable instructions to its bankers to hold the documents of title relating to the cargo to the order of Glencore. At one stage in these proceedings there was a dispute between the parties as to whether Glencore acquired title to the oil purchased under these back-to-back contracts, but by the time the trial ended it was common ground that title to the oil brought into Fujairah under these arrangements had vested in Glencore by the time the carrying vessel arrived in Fujairah.
8. Parcels of oil held by MTI in storage for the account of the joint venture were transferred by Glencore to MTI under individual in-tank transfers known as "ITT contracts". These took the form of individual contracts of sale which provided for payment by letter of credit and for title to the oil to which they related to be transferred to MTI by the issue of a stock transfer certificate. It will be necessary to consider the form of these ITT contracts in some detail at a later stage, but for present purposes it is sufficient to draw attention to the following features: first, they provided for payment 45 (later 55) days after the stock transfer date against presentation of an invoice and stock transfer certificate; secondly, the stock transfer was to take place on receipt by Glencore of a letter of credit; but, thirdly, a specific date was identified as the date on which the stock transfer was deemed to have taken place for payment purposes; finally, they provided for title to pass to MTI on delivery of the stock transfer certificate.
9. Although there was no joint venture vehicle as such, it was convenient for both parties to treat oil brought into Fujairah under these arrangements as if it belonged to the joint venture. Both Glencore and MTI kept records of the quantities of oil held for the account of the joint venture and periodically exchanged information for the purposes of reconciling their positions and calculating the amounts due from one to the other. In due course I shall have to refer in some detail to these records, but it is a striking fact that throughout the life of JV1 and JV2 Glencore and MTI both recorded oil as leaving the joint venture inventory and entering the MTI inventory on the date identified in the ITT contracts as the date on which stock transfer was deemed to take place for payment purposes rather than on the date of issue of the stock transfer certificate which could be, and often was, many days later.
10. JV1 came into operation at the end of 1994 or early 1995 as a result of negotiations between Mr. Heuzé and Mr. Kilakos. At the trial there was a dispute between the parties as to the precise circumstances in which that came about and the terms which applied to it, but it appears to have grown quite naturally out of the business which the parties had been doing with each other over the preceding year and it appears to have operated without any apparent difficulty until June 1996 when further negotiations between Mr. Heuzé and Mr. Kilakos led to the modified arrangements of JV2. Again, there is a dispute about the precise nature and terms of this arrangement to which I shall return later.
11. The parties continued to operate under the JV2 arrangements until February 1998. Although Glencore was responsible for providing the finance to enable cargoes to be purchased from prime suppliers, supply arrangements were largely in the hands of MTI. That was inevitable since under JV2 MTI undertook to purchase from Glencore all the oil brought into Fujairah under these arrangements and bore the complete responsibility for re-selling it into the market. During the latter part of 1997 the volume of oil brought into Fujairah under JV2 increased markedly. This increase in stock levels coincided in part with a substantial fall in the fuel oil market, which was deeper and longer-lasting than any previous downturn most traders could recall. By the end of the year the price for 380cst fuel oil, a staple commodity in this market, had fallen from around $100 per ton to around $50 per ton. This decline in the market placed MTI under severe financial pressure because it had committed itself to buy the components required for producing blended fuel oil at prices which it could not cover by sales into the bunker market. By the end of January 1998 it had exhausted its credit lines with its banks and could no longer open letters of credit to pay Glencore for oil drawn from the joint venture stock. During the first week of February 1998 MTI effectively collapsed.
12. On 7^th February 1998 a meeting took place in Athens between representatives of Glencore and MTI at which it was established that large quantities of oil which Glencore had brought into Fujairah under the JV2 arrangements, but which had not yet been the subject of a stock transfer certificate formally transferring title to MTI, were no longer in MTI's possession. Some of that oil had been sold to MTI under ITT contracts, but had not been paid for nor had any letter of credit been opened in respect of the price. The rest had not been made the subject of an ITT contract at all. Glencore maintained that during the preceding months MTI had been dishonestly selling oil drawn without authority from the joint venture stock as a means of keeping itself afloat. In a letter written to its insurers shortly after that meeting it described the loss of its oil as due to theft on the part of MTI and a claim was made under the policy on that basis. At the meeting on 7^th February Mr. Kilakos and his son John signed an agreement on behalf of MTI which recorded that MTI could not account for about 1.4 million tons of oil which it was required to hold to Glencore's order. The agreement also gave Glencore the right to control and monitor MTI's assets and its various business operations. Mr. Kilakos maintained that he had signed that agreement under duress and it will therefore be necessary to say something more about the events of 7^th February in due course.

The parties' contentions

13. Glencore's primary contention in this litigation is that as between itself and MTI it retained title to all the oil remaining in store at Fujairah at the time when MTI collapsed. Mr. Alistair Schaff Q.C. submitted on its behalf that although individual cargoes lost their identity as a result of being commingled in common storage, Glencore became an owner in common of all the oil of the relevant grades in proportion to the amount of oil it had contributed to the common bulk after taking into account the quantities in which property had passed to MTI under ITT contracts. It is Glencore's case that MTI had exhausted the whole of the interest it may have had in each of the different grades of oil so that, whatever may be the position as regards other oil claimants, MTI no longer held a proprietary interest in any of the oil still in stock. However, the extent to which any party, including MTI, may be entitled to assert a proprietary interest in the remaining stock is not a matter which falls for decision in this Phase.
14. The bedrock of Mr. Schaff's submission was that the ITT contracts regulated the passing of title to oil held by MTI in accordance with their terms. It had become common ground that Glencore was the owner of the oil at the time it was delivered into the floating storage and title passed to MTI only under and in accordance with the terms of the ITT contracts. Property in the oil therefore remained in Glencore until it issued a stock transfer certificate certifying that title had passed to MTI. Insofar as delivery occurred at an earlier date, that did not affect title to the oil.
15. Mr. Michael Crane Q.C. on behalf of MTI submitted that from as early as the latter part of 1994 the ITT contracts by themselves ceased to reflect fully the agreement of the parties as it had by then developed through the course of business. The original terms of the parties' relationship had been modified through the operation of the JV1 arrangements and these had been further modified when the JV2 arrangements had been agreed in June 1996. He submitted that title in the oil delivered into Fujairah by Glencore under both JV1 and JV2 passed by operation of law to MTI, but that even if that were wrong, MTI was by agreement allowed to dispose freely of the oil from the moment it was discharged into the floating storage. Accordingly, property passed by agreement from Glencore to MTI at that point.
16. The 'third way' for which the insurers contended fell between these two extremes. Mr. Jeremy Cooke Q.C. on their behalf submitted that in the course of doing business with each other Glencore and MTI had reached an agreement, notwithstanding the terms of the ITT contracts, that oil was to be released to MTI at what was called for convenience the "ITT date", that is, the date identified in the ITT contract when the stock transfer was deemed to take place for payment purposes. That was the date on which oil sold under an ITT contract was recorded by both sides as moving from the joint venture inventory to the MTI inventory and he submitted that the release of oil in this way was intended to give MTI the right to dispose of it in any way it thought fit. Accordingly, title passed to MTI at the moment of release, or, if it did not, Glencore consented to MTI's dealing with the oil in the ordinary course of business so that any subsequent disposal of the oil operated to pass title to it.

Secured transactions

17. The concept of a secured transaction lay at the heart of the evidence given by the principal witnesses for Glencore and the submissions made on its behalf by Mr. Schaff. A clear distinction can be drawn in commercial terms between an agreement of the kind suggested by Glencore and agreements of the kind suggested by both MTI and the insurers. The former is a secured transaction in the sense that title to the goods does not pass to the buyer until a letter of credit securing payment of the price has been opened in favour of the seller. Transactions of the latter kind are essentially unsecured in that the release of the goods and the passing of property may occur before the opening of a letter of credit. Under transactions of this kind Glencore would be exposed to the risk of MTI's failing to pay for the goods.
18. All the witnesses who had experience of the international oil trade agreed that it is usual for those engaged in that trade to do business on letter of credit terms unless they are selling to one of the major oil companies. Glencore is a large and very experienced trading house and it would be surprising, therefore, if at the beginning of its relationship with MTI, a small and relatively unknown company, it had been willing to sell oil on any other terms. However, as the relationship between the two companies developed it is by no means impossible that a sufficient degree of trust was generated to lead it to trade on less restrictive terms. It is necessary therefore to examine the whole of the course of dealing between the two parties to see to what extent, if at all, it developed in the way MTI or the insurers suggest.

The early transactions

19. The first transactions between these two parties occurred in 1992 and involved the exchange of products. In the spring of 1992 Glencore had obtained a contract to supply fuel oil to Pakistan which it intended to fulfil with fuel oil purchased from a refinery in Bahrain. In the event there was some doubt whether the cargo which had been loaded in Bahrain complied in all respects with the contract of sale and Glencore therefore entered into a contract with MTI for the exchange of part of the cargo for a fuel oil of higher quality. These were carried out by ship to ship transfer at Fujairah and were successful in bringing the overall specification of the cargo up to the level required by the contract. This was followed by a second transaction of a similar kind about a week later.
20. In May 1992 Glencore began to trade with MTI in a more usual way, selling cargoes of Bahrain fuel oil to MTI on f.o.b. terms and buying from MTI c.&.f. Port Qasim cargoes of fuel oil whose specification corresponded to that of its contract with its Pakistani buyer. In each case payment was made by letter of credit. These transactions were negotiated between Mr. Heuzé and Mr. Griffin and were followed in August 1992 by the first joint venture arrangements under which Glencore and MTI agreed to share in the disposal of a cargo of fuel oil belonging to Glencore on board the vessel World Amber. This was again negotiated between Mr. Heuzé and Mr. Griffin. This first joint venture, in common with those that succeeded it, was set up essentially as an accounting exercise under which oil was charged to the enterprise by Glencore at an agreed price and subsequently sold either to a third party buyer or, at a later date, to MTI itself. The profit or loss on the transaction, calculated after taking account of the price to the joint venture and the costs associated with its storage and marketing, was split equally between Glencore and MTI. There was no joint venture vehicle at any stage and title to the oil remained in Glencore until it was sold to the purchaser. A second joint venture was concluded between Mr. Heuzé and Mr. Griffin on similar terms in relation to the cargo carried on the Obo Vega. Neither of these cargoes were held in storage at Fujairah, although the cargo on the Obo Vega was in fact sold by Glencore to MTI and discharged into MTI's facility. Glencore also sold a parcel of light cycle oil loaded on to the Obo Vega in Malta outside the terms of the joint venture.

The Handy Grace agreements

21. There appear to have been no further transactions between Glencore and MTI until June 1993 when Glencore was looking for a way to dispose of a cargo of Iranian fuel oil on a vessel called the Handy Grace for which it was unable to find an immediate buyer. Mr. Heuzé approached Mr. Griffin with a view to arranging the storage of the cargo by MTI. The outcome was an agreement recorded in a telex message of 16^th June 1993 from Glencore to MTI under which MTI agreed to store the cargo for a period of 20 days for a fee of US$1.25 per metric ton. It was expressly agreed that the oil could be held in commingled storage, although it is implicit in the fact that MTI undertook to endeavour to redeliver the same quantity and quality of oil that it would be commingled only with oil of substantially the same kind. The agreement expressly provided that title and risk in the oil should always remain with Glencore. The cargo was discharged into the Metrotank on 18^th-19^th June 1993.
22. In August 1993 Mr. Heuzé entered into an agreement with Mr. Griffin for the sale by Glencore to MTI by in-tank transfer of a parcel of 15,000 tons of high sulphur fuel oil from the Handy Grace cargo. This contract is of significance because it was the first occasion on which Glencore had sold oil to MTI by in-tank transfer. The terms of that contract were set out in a telex message from Glencore to MTI dated 19^th August 1993 which was copied to Mr. Griffin in the United States. They provided for delivery to be made by stock transfer on 18^th August and for payment by letter of credit 45 days from the stock transfer date against presentation of an invoice and stock transfer certificate. The contract also provided that risk should pass from seller to the buyer on 18^th August and that title should pass on exchange of the title document. MTI entered the oil in its own inventory as a delivery made on 19^th August.
23. Mr. Heuzé and Mr. Griffin entered into a second agreement in substantially the same terms a few months later as confirmed by a telex message dated 19^th November. The quantity sold on this occasion was 36,547.386 tons and delivery was to be made by stock transfer on 9^th November. Neither Mr. Heuzé nor Mr. Griffin was able to recall why the delivery date had been backdated in that way.
24. In January 1994 Glencore sold MTI 21,547.386 metric tons of fuel oil representing the remainder of the Handy Grace cargo. The contract was confirmed by a telex message dated 17^th January 1994 which sets out in detail the terms of the transaction. These were generally very similar to those of the first and second in-tank transfers, but provided for delivery to take place by stock transfer on the date of receipt by Glencore of a letter of credit. The contract provided for risk to pass from seller to buyer on the stock transfer date and for title to pass on exchange of title documents. An additional 'special condition' was incorporated into this contract providing that stock transfer was to take place upon receipt of an irrevocable letter of credit.

The Kulab agreements

25. In early December 1993 Glencore entered into another agreement with MTI for the storage of oil, this time in respect of two parcels of about 8,500 tons of Iranian straight run fuel oil to be delivered to Fujairah in barges. As in the case of the Handy Grace storage agreement, it was expressly agreed that the oil could be commingled in storage and that the goods should remain the property of Glencore, but on this occasion the agreement expressly contemplated that the oil would in due course be sold to MTI. The barge Kulab was used to deliver the oil into Fujairah and accordingly this arrangement has become known as the Kulab storage agreement. The arrangement was recorded in a telex message dated 2^nd  December 1993.
26. At the same time as it entered into the Kulab storage agreement Glencore entered into an agreement to sell MTI by in-tank transfer about 17,000 tons of fuel oil in two separate parcels representing the two parcels discharged from the Kulab into the Metrotank. The ITT contract, which was contained in a separate telex message dated 2^nd December 1993, provided for delivery of the two parcels to take place by stock transfer between 15^th and 30^th January and between 25^th January and 10^th February 1994 respectively. Stock transfer was to be effected by the buyers' making a declaration of Platt's quotations for pricing purposes. Payment was to be made by letter of credit against invoice and stock transfer certificate. In this case, however, the contract specifically provided that risk and title should pass to the buyer on the stock transfer date.

The Mount Athos agreements

27. At about the same time Glencore and MTI entered into three related agreements under which Glencore agreed to purchase from MTI four or five cargoes of Iranian straight run fuel oil which MTI had already agreed to buy from NIOC. These cargoes were to be stored by MTI and subsequently disposed of under a joint venture arrangement. The intention behind this transaction was to take advantage of unusual market conditions to execute what is known as a 'contango play'. Broadly speaking, contango conditions exist when the forward price of a commodity is significantly higher than the price for prompt delivery. Depending on the difference between the prices it may be possible to buy goods for prompt delivery, store them for a period and sell them at a later date at a profit after allowing for the cost of the original purchase and all the costs associated with keeping them in storage in the interim. By making use of recognised forms of hedging transactions it is possible to insulate the transaction from market movements and establish a fixed profit on the overall transaction, provided the goods are available for prompt delivery on the agreed date. In order to take advantage of the opportunity to make a certain profit, therefore, it is necessary to hold the goods in storage in order to ensure that they are available for delivery on the due date.
28. The three agreements in question were all negotiated between Mr. Heuzé and Mr. Griffin and were confirmed in a series of telex messages all dated 17^th December 1993. The first was an agreement under which Glencore agreed to buy the cargoes of fuel oil from MTI on discharge into floating storage at Fujairah. The second agreement was a storage agreement under which MTI agreed to hold the oil in segregated storage between January and the end of March 1994. In order to enable MTI to store these cargoes separately from other oil and to ensure their integrity it was agreed that they would be held on board a vessel specially chartered for that purpose, the Mount Athos. The storage fee payable by Glencore represented half of the cost of chartering and operating the Mount Athos as a storage vessel. The storage agreement also provided that Glencore would sell at least half of the oil to MTI on letter of credit terms. This provided a cross-reference to the third agreement which was described as a joint venture agreement. Under it MTI agreed to buy from Glencore by way of stock transfer no later than 31^st March 1994 half of the oil delivered into storage at a price which reflected the full cost of acquisition and storage. The joint venture agreement specifically provided that risk and title to the oil would pass to MTI only on issue of a stock transfer certificate and that Glencore would issue a stock transfer certificate only on receipt of an acceptable letter of credit. Profits from the sale of the oil were to be shared equally. Although they agreed to co-operate in marketing the oil to maximise their profits, each party was responsible for its own hedging arrangements and for the disposal of its own share of the oil. This is emphasised by the fact that Glencore was given an option to sell all or part of its share of the oil still on board the Mount Athos in February 1994 to MTI at a market-related price.

The gasoil cargoes

29. At this early stage Glencore and MTI continued to enter into 50/50 joint venture agreements relating to specific cargoes of oil. In January 1994 Mr. Heuzé and Mr. Griffin entered into such an arrangement in relation to a cargo of gasoil shipped on the vessel Iter. The cargo was purchased with a view to its sale in the bunker market in Fujairah. MTI bought the cargo from the prime supplier and sold it to Glencore on back-to-back terms. The cargo was stored by MTI pending its disposal and it was agreed between them that Glencore would sell any part of its share which remained in stock at the end of April to MTI at a market-related price. In this case, although each party was responsible for the sale of its own share of the oil, profits and losses were to be shared between them. In May 1994 Glencore and MTI made a similar arrangement in relation to a cargo of gasoil carried on the vessel Macle.

The May 1994 storage agreement

30. In May 1994 Glencore and MTI entered into a new storage agreement to replace the one made on 17^th December 1993. Again, it was negotiated between Mr. Heuzé and Mr. Griffin. It was described as a storage agreement for cargoes of Iranian atmospheric straight run fuel oil and was expressed to run from 18^th March 1994 until terminated by mutual agreement on 30 days' notice. Under the agreement MTI agreed to store cargoes of Iranian straight run fuel oil on board the Mount Athos or the Metrotank or other vessels at a fee of 6 cents per ton per day for oil already in store at 16^th March and $1.25 per ton for the first 20 days and 5 cents per ton per day thereafter in respect of oil delivered into storage after 16^th March.
31. The May 1994 storage agreement was necessitated in part by the fact that the Mount Athos, which had been on charter to MTI, was about to leave Fujairah. The parties retained the previous provisions which required MTI to ensure that the same quantity as delivered by Glencore should at all times be held to its order and for the master of the relevant storage vessel to confirm that he held the oil on board his vessel to the order of Glencore alone, but on this occasion the contract no longer contained a provision entitling Glencore to sell its share of the oil in store to MTI. By its terms the May 1994 storage agreement applied only to Iranian straight run fuel oil. During 1994 and thereafter parcels of Iranian straight run fuel oil were brought into Fujairah for storage from time to time, but as time went on other types of oil were also brought into storage. The May 1994 storage agreement was never formally terminated, nor was it ever expressly agreed that it should apply to the other types of cargoes which were brought into Fujairah during the following months, but no other terms were agreed between the parties to apply to cargoes of other kinds. In these circumstances there was a dispute as to whether the May 1994 storage agreement had been informally adopted in relation to a wider range of cargoes, or whether the storage of these other cargoes was unregulated by any agreement. This is a question to which I shall have to return in due course.

The succeeding months

32. Between May and early October 1994 several cargoes of Iranian straight run fuel oil of 180cst and 280cst grades were brought into storage at Fujairah, all of which were held by MTI and disposed of under the terms agreed in December 1993 and modified in May 1994. From the early part of October 1994 the range of cargoes was extended to include for the first time Iranian fuel oil of 380cst grade. The very first cargo of 380cst oil brought to Fujairah was carried on the vessel Guglielmo Effe, but in the event it was carried on to its destination without being discharged and did not go into storage at all. The first cargo of 380cst fuel oil discharged into storage was carried on the Jaguar. It is recorded as entering the joint venture inventory in mid-November.

The state of the parties' relationship in November 1994

33. The history of these early transactions is of some importance in analysing the development of the relationship between Glencore and MTI. Mr. Schaff submitted that by the beginning of November 1994 the basis on which Glencore and MTI were prepared to do business together as joint venturers had been clearly established. It was derived from the agreements made in December 1993 as modified by the May 1994 storage agreement, all of which were intended to remain in place as a general framework under which the parties did business, and involved a recognition on the part of MTI that oil brought into Fujairah for storage and disposal under the joint venture belonged to Glencore at the time of its delivery into storage. Thereafter it remained Glencore's property until it was sold to MTI under the terms of an ITT contract. A standard form of ITT contract had been established under which oil was delivered to MTI against opening of a letter of credit covering payment of the price and title passed on delivery of a stock transfer certificate. The ITT contracts were, therefore, secured transactions and both parties were well aware that Glencore was not willing to trade with MTI except on secured terms. In those circumstances, he submitted, the onus was on MTI to show when and how the parties agreed to depart from that arrangement and to do business on unsecured terms.
34. Mr. Crane put forward two arguments in response. The first concerned the effect of commingling in common storage. He submitted that in a case where goods are deposited with a warehouseman for storage in common with other goods of the same description, the owner only acquires title as an owner in common to the bulk to which his goods made a physical contribution. He submitted that three consequences flow from this. The first is that the original owner cannot assert a proprietary interest in the whole of the stock of goods of the same description held by the warehouseman; his claim only lies against the particular bulk to which his goods contributed. The second is that if it has become impossible to identify the bulk to which his goods contributed, the original owner is precluded from asserting a proprietary claim to any of the goods held by the bailee. The third is that unless the original owner can identify the bulk to which his goods contributed, property in them passes on delivery to the warehouseman. Mr. Crane submitted that it was impossible in this case to identify at any stage the specific bulk to which Glencore's goods contributed because no records were kept of a kind which would enable that to be done. As a result, from the earliest stages property in the oil passed to MTI as soon as it was delivered into common storage at Fujairah, regardless of the parties' intentions. This is a submission to which I shall return because it can most conveniently be considered in the light of my conclusions on the nature of the parties' agreement.
35. Mr. Crane's second submission was that the transactions between Glencore and MTI which took place between April 1992 and November 1994 did not establish any clear basis of trading as suggested by Mr. Schaff. They were of a variety of different kinds and were carried out on a variety of different terms. They were not all secured transactions and do not bear out the conclusion that the parties had established a clear basis for their future dealings. One cannot therefore approach the next stage of the parties' relationship on the basis that all subsequent transactions were intended to be of a secured nature unless there was a clear agreement to the contrary.
36. Sales by in-tank transfer are not uncommon in the oil industry, but one particular feature of the transfers between Glencore and MTI which may have made the contracts more difficult to structure satisfactorily was the fact that the oil in question was already in the possession of the buyer, MTI, at the time of the contract. (I can put on one side those cases in which the master of the storage vessel issued a telex confirmation direct to Glencore that he was holding oil to its order. Despite the terms of the storage agreements, very few telexes of that kind were provided and only in the case of the cargo on the Mount Athos did the telex evidence a relationship of bailment between the owners of the vessel and Glencore.) Despite that, it is a characteristic of almost all the ITT contracts that delivery is expressed to occur on the stock transfer date, whereas title is to pass later against delivery of a stock transfer certificate. Moreover, in some cases the stock transfer date was earlier than the date of the ITT contract itself. For example, under the second Handy Grace ITT contract dated 19^th November 1993 the stock transfer date was 9^th November 1993.
37. Almost all the ITT contracts in this case were drafted by a member of Glencore's Operations Department, Mr. David Hawkins. The function of the Operations Department is to provide administrative support to the commodity traders such as Mr. Heuzé by drafting and reviewing contracts, letters of credit and other documents required to enable contracts entered into by Glencore to be performed. Mr. Hawkins had no formal training; he learned his skills on the job and by the time he became involved in the business with MTI he had had about two years' experience dealing with operations relating to oil cargoes. I mention this because I think it explains in part why the documentation in this case takes the form it does. When Mr. Hawkins was asked to draw up the paperwork to cover a new piece of business his reaction, perhaps not unnaturally, was to take as his starting point the paperwork relating to a previous contract of a similar kind and adapt it as seemed appropriate. No doubt this approach was generally satisfactory, but it obviously entailed the risk that provisions might be carried over from an earlier document which were not wholly appropriate to the contract in hand.
38. When Mr. Hawkins began work in the Operations Department he was supporting a number of fuel oil traders, including Mr. Heuzé. He was not involved in drafting the Handy Grace agreements, but he was responsible for drafting the documents relating to the Kulab cargoes and virtually all the documents which were produced by Glencore relating to its subsequent transactions with MTI. As far as the Kulab parcels were concerned, Mr. Hawkins was told that there was to be a storage agreement similar to that relating to the Handy Grace cargo and that the oil in question had been sold to MTI in two parcels by in-tank transfer during January and February 1994. It is hardly surprising, therefore, that when drawing up the contract for the Kulab parcels he used the Handy Grace documents as his starting point.
39. Nothing in Mr. Hawkins' evidence led me to think, however, that he gave any very careful consideration to the distinction between the transfer of possession involved in delivery and the transfer of risk on the one hand and the passing of title on the other. Certainly there is nothing to indicate that he was aware of any commercial reasons for structuring the ITT contract in that way. As the trader who had made the agreement to sell the Handy Grace cargo Mr. Heuzé might be expected to have been more alive to this distinction. He accepted that he probably saw most of these early ITT contracts, but his response on being asked why they took this form was simply to say that they were "not logical" since both title and risk should have passed against delivery of a stock transfer certificate. In the end none of Glencore's witnesses was able to give a satisfactory explanation of the commercial purpose behind the separation of the transfer of possession and title which one finds in what became the established form of ITT contract. Once it had appeared in the Handy Grace ITT contracts, however, it found its way in broadly similar terms into all subsequent ITT contracts.
40. Another feature common to all the ITT contracts entered into during this early period is the requirement for payment by letter of credit. All these transactions were negotiated between Mr. Heuzé and Mr. Griffin. Mr. Griffin accepted that at the time of the first Handy Grace ITT contract he was aware that Glencore was only prepared to sell oil to MTI on letter of credit terms and Mr. Kilakos himself recognised as much. Mr. Griffin also accepted that in the ordinary way a seller who demanded payment by letter of credit would not intend the buyer to have the right to dispose of the oil before the letter of credit had been opened. Although Mr. Heuzé's contacts with Mr. Kilakos were becoming more frequent during 1994, Mr. Griffin continued to be the primary point of contact for negotiations at that time. He accepted that throughout the summer of 1994 he understood that it was Glencore's intention that the oil sold to MTI under ITT contracts should only be at MTI's disposal once it had opened a letter of credit. His view was that both Mr. Kilakos and Mrs. Gene understood that to be the position and certainly Mrs. Gene's evidence tended to bear that out.
41. It is fair to say that on this issue Mr. Kilakos' evidence was somewhat confused: one moment he said that he understood MTI to be entitled to make use of the Handy Grace cargo from the moment it was discharged into storage, with MTI merely being obliged to pay Glencore the market value of the oil at the time when it sought redelivery; the next he said that MTI had to return oil in kind to Glencore if that was what Glencore wanted. Having regard to the terms of the Handy Grace storage agreement (which I am confident were brought to his attention) I find it difficult to accept that Mr. Kilakos really thought that MTI had the right to dispose of the oil from the moment it was placed in storage and certainly he offered no reason for holding that view. I think much of the confusion resulted from the fact that he was aware that MTI was entitled to commingle the oil in storage with oil which it was holding for its own purposes and was equally aware that its original identity was inevitably lost in the commingling process. It is inherent in the agreement that MTI was not obliged to redeliver the very same molecules of oil as it had received from Glencore and so in one sense it might be said that it was entitled to use oil from the Handy Grace as soon as it was put into storage. At the same time, however, it was obliged to keep an equivalent quantity of oil of the same grade available for redelivery to Glencore on demand. Mr. Kilakos' evidence in relation to delivery and the passing of title under these early ITT contracts was equally unenlightening. He does not have legal training and cannot be criticised for failing to appreciate the legal distinctions involved, but in common with the other witnesses he was unable to say what commercial purposes might be served by structuring the ITT contracts in this way. I am satisfied that at that stage none of those involved had given any serious consideration to the implications of framing these contracts in that way.
42. The three Handy Grace ITT contracts are certainly capable of being construed in a manner consistent with that understanding. Although they provide for delivery to take place and for risk to pass to MTI on the stock transfer date, the express provision for title to pass on exchange of title document reflects the understanding of the parties that the oil was not to be at the disposal of MTI until that time. Each of these can therefore be regarded as a secured transaction and indeed the position is even stronger in the case of the third ITT contract which provided for the stock transfer to take place on receipt of a letter of credit.
43. The same cannot be said of the Kulab ITT contracts, however, which in their original form provided that risk and title was to pass on the stock transfer date established by MTI's declaration of dates for pricing purposes. Although the contract provided for payment by letter of credit and for payment to be made against a stock transfer certificate, it would be difficult to say that the passing of title was deferred until after the letter of credit had been opened. Mr. Crane submitted that this showed a willingness on the part of Glencore at this early stage to sell to MTI on 'open account' (i.e. unsecured) terms, but I find that difficult to accept for a number of reasons. In the first place, although I do not discount the possibility that Glencore may in due course have been prepared to trade with MTI on open account terms, this transaction took place quite early in the parties' trading relationship and I think it would have been surprising if Glencore had been willing to forego the benefits of a secured transaction at that stage. Secondly, none of those involved seem to have realised at the time that that was the effect of the contract. In this respect I think it is telling that Mr. Griffin, whom I found to be generally a very open and reliable witness, did not identify this as an example of the parties' consciously agreeing to trade on open account terms. Thirdly, the way in which these two cargoes were ultimately disposed of points to the same conclusion. The first cargo of just under 8,000 tons was discharged into the vessel Jaguar on 11^th December 1993. An analysis of a sample drawn from the vessel's tanks immediately prior to discharge indicated that the flash point was too low. This led to a renegotiation of the contracts and a spot sale by Glencore to MTI of this first parcel at a substantially reduced price. Title passed on discharge at Fujairah and the price was payable by telegraphic transfer 30 days after notice of readiness. This was clearly not a secured transaction, but the fact that Mr. Griffin obtained this revision tends to support the conclusion that neither trader thought that the original transaction was unsecured.
44. The second Kulab cargo was unaffected and should therefore have been handled under the terms originally agreed. In the event, however, no letter of credit was opened and a stock transfer certificate was sent by telex together with an invoice payable at 45 days from notice of readiness. There is nothing in the documents to explain this departure from the agreed terms, but Mr. Heuzé said in evidence that Mr. Griffin had quite properly taken advantage of the problem with the first cargo to obtain Glencore's agreement to trading both parcels on open account terms. Mr. Crane suggested that this explanation was disingenuous and that Glencore did not require a letter of credit because the transaction was always intended to be handled on open account terms. I am unable to accept that. If it had been intended from the outset that these sales were to be unsecured it is difficult to see why Glencore stipulated for payment by letter of credit in the first place or why Mr. Griffin (who would surely have appreciated the fact) did not object to a confirmation telex which provided for it. Mr. Griffin's evidence on this point does not suggest that he understood that the original Kulab transaction was intended to be unsecured, but it does support Mr. Heuzé's evidence that open account terms were introduced as part of the renegotiation. What remains unexplained, however, is exactly when and under what circumstances open account terms were agreed in relation to the second cargo. The terms agreed in the exchange of telex messages following the discovery that the first cargo was off-specification did not affect the second cargo which continued to be subject to the original terms. At some stage there must have been an agreement to dispense with the letter of credit, but when and how that came about remains unclear. At all events, I am confident that it was the result of specific agreement between the parties and not simply the result of Glencore's realising that the transaction was effectively unsecured.

Sales of oil on board the Mount Athos

45. On 8^th April 1994 Glencore agreed to sell to MTI two parcels of fuel oil on board the Mount Athos totalling 15,000 tons. ITT contracts relating to these parcels were issued on the same day, each of which provided for payment by letter of credit and for title to pass on exchange of title document. The ITT contracts provided that delivery should take place by stock transfer from Glencore's cargo on board the Mount Athos and on the same day Glencore sent instructions to the master authorising the release of the oil to MTI. On 11^th April Mr. Griffin reported the transaction to MTI's Athens office and asked for the letter of credit to be opened as soon as possible in order to facilitate the release of a further parcel of oil. A letter of credit covering the payment of both parcels was in fact opened the next day. On 14^th April Glencore issued a stock transfer certificate certifying that title to 15,000 tons of fuel oil on the Mount Athos had been transferred to MTI.
46. On 11^th April Glencore agreed to sell MTI a further parcel of 15,000 tons of oil on the Mount Athos to MTI. Following the prompt opening of the letter of credit covering the first two parcels Glencore sent a message to the master of the vessel on 13^th April authorising him to release the oil to MTI immediately. On the same day Glencore issued an ITT contract covering the sale of the oil which again provided for payment by letter of credit and for title to pass on exchange of title document. The letter of credit was opened on 18^th April and Glencore issued its stock transfer certificate on 19^th April.
47. A third transaction of this kind soon followed. On 22^nd April Glencore agreed to sell MTI a parcel of 30,000 tons of fuel oil on board the Mount Athos and a few days later a further 45.540 tons was added to that as representing the balance of a cargo that had been discharged from another vessel, the Pacific Jade. These parcels were released to MTI in the same way on 22^nd and 25^th April. On 25^th April Glencore issued an ITT contract in respect of 30,045.540 tons of fuel oil in substantially the same terms as the two previous contracts. On this occasion, however, delivery was said to have been made by stock transfer on 22^nd April. The letter of credit was opened on 27^th April and a stock transfer certificate was issued on 3^rd May. One can see from the letter of credit and Glencore's invoice that the oil had already been loaded onto another vessel, the Neptune Lynx, under a bill of lading dated 28^th April.
48. Both Mr. Kilakos and Mrs. Gene suggested in their statements that these transactions were typical of the way in which the parties did business at that time and, according to Mr. Kilakos, were evidence of the fact that MTI was entitled to dispose of oil as soon as it had been brought into storage under joint venture arrangements. I am quite unable to accept that, and indeed it is fair to say that in cross-examination Mrs. Gene accepted that she was aware that in general Glencore was only willing to do business with MTI at that time on secured terms. Mr. Crane did not seek to suggest that these transactions showed that MTI was intended to have free access to oil in storage at this time, but he did submit that they marked the beginning of a developing relationship of trust under which Glencore was increasingly willing to release oil to MTI in advance of the opening of a letter of credit.
49. Mr. Heuzé accepted in cross-examination that in each of these cases Glencore had been willing to give MTI unrestricted access to the oil before a letter of credit had been issued and it follows, as indeed he accepted, that in these instances Glencore had been willing to deal with MTI on an unsecured basis. His explanation, however, was that a sufficient degree of trust had by then been generated between himself and Mr. Griffin.
50. I have little doubt that as their relationship developed, and particularly as Mr. Heuzé and Mr. Kilakos, and Mr. Hawkins and Mrs. Gene, became better acquainted, the degree of trust between Glencore and MTI grew. Mr. Heuzé accepted that and throughout his evidence was willing to accept that until the business collapsed in 1998 MTI had always performed in accordance with its contracts. However, I am unable to accept that the release of oil by Glencore in April 1994 demonstrated, or was understood by MTI to indicate, a general willingness on its part to trade with MTI on unsecured terms. Mr. Heuzé's explanation, broadly supported by Mr. Griffin, was that MTI had requested these prompt releases of oil as something of a favour and that Glencore had been willing to accommodate MTI to this limited extent, transaction by transaction. I do not think one can infer from that that there was a general willingness to trade on unsecured terms at that time, and certainly that was not Mr. Griffin's understanding. No further releases of this kind occurred and it is relevant to note that during the next few weeks one can find in the context of sales by in-tank transfer a number of messages from Glencore to MTI confirming that stock transfer would take place on MTI's providing financial security.

The standard form of ITT contract

51. The early ITT contracts differed in their terms, but it was common ground that by June 1994 a form had emerged which thereafter remained substantially unchanged. It provides for delivery by stock transfer and for payment by letter of credit 45 days after stock transfer date against presentation of invoice and stock transfer certificate. Risk was to pass from seller to buyer on the date of the stock transfer and title on delivery of the stock transfer certificate. A 'special condition' provided that stock transfer would take place on receipt of a letter of credit but that for payment purposes the stock transfer date would be deemed to have taken place on a stated date. This form of contract is entirely consistent with the understanding of Mr. Heuzé and Mr. Griffin (and also of Mrs. Gene) that at this stage Glencore intended to deal with MTI on secured terms subject to any specific agreement to the contrary.
52. Taken as a whole I think the evidence supports Mr. Schaff's submission that by the autumn of 1994 there was an established pattern of dealing between Glencore and MTI under which it was well understood that Glencore would sell oil to MTI only on secured terms and, in the case of in-tank transfers, on the terms of what had by then become the standard form of ITT contract.

Entry into the bunker market

53. Mr. Crane submitted that a fundamental change occurred in the commercial nature of the parties' relationship towards the end of 1994 when the decision was taken to enter the bunker market in Fujairah. Glencore and MTI had previously co-operated in the purchase and sale of various cargoes on a joint venture basis, but although these joint ventures had many features in common, each had been a separate transaction subject to specific agreement. Moreover, he submitted, the agreements which had been put in place between December 1993 and May 1994 to govern the joint venture relating to the Iranian straight run fuel oil cargoes were never intended to apply to other types of cargo or other kinds of commercial activity. The parties may have treated them as applicable to a few additional cargoes of Iranian fuel oil which were brought into Fujairah during the summer of 1994, but they did not by their terms cover cargoes of 380cst bunker fuel and were never intended to do so. No terms for this new business were specifically agreed and therefore in order to ascertain the terms on which the parties did business it is necessary, he submitted, to examine the way in which that business was conducted.
54. Before turning to consider whether the parties' relationship took a new commercial direction at the end of 1994 it is necessary to describe the background in a little more detail. As soon as the parties embarked on any co-operative ventures a need arose on both sides to keep a record of the transactions and to reconcile the parties' positions from time to time. In about January 1994 Mr. Hawkins began keeping two sets of records: an inventory of the various products held by MTI in storage for Glencore and a financial record headed 'Value of Cargoes FO stored aboard MT Mount Athos and Metrotank bought/sold by Marc Rich [Glencore]'. At the outset he recorded oil as leaving the inventory under a sale to MTI on what became the established ITT contract terms on the date when the stock transfer certificate was issued. However, he recorded the parcels on board the Mount Athos which were the subject of the special releases in April 1994 as leaving inventory on the date of the release and the next three parcels from the Mount Athos which were the subject of instructions to the master for release on 6^th, 10^th and 15^th May respectively he recorded as leaving the inventory on those dates. In the case of the following in-tank transfers Mr. Hawkins adopted the stock transfer date as the date on which oil left the inventory rather than the date on which the letter of credit was opened or the date on which the stock transfer certificate was issued. However, the ITT contracts in these cases did not identify a date on which stock transfer should be deemed to have occurred for payment purposes (as they did from June 1994), although they did contain a clause providing that stock transfer would take place on opening of a letter of credit. What appears to have happened is that a stock transfer date was subsequently agreed between the parties which in some cases was a few days before the letter of credit was opened. That date was then recorded as the date on which the oil left the inventory.
55. During 1994 and the first half of 1995 the parties held reconciliation meetings at roughly quarterly intervals to agree their figures and to establish profits and losses under joint venture arrangements. At one of the early meetings held in London, probably in June 1994, Mr. Griffin gave Mr. Hawkins a computer disk containing a copy of the spreadsheet which he himself was using to record movements of oil brought to Fujairah under the parties' joint venture arrangements. Thereafter Mr. Hawkins kept his inventory records on the spreadsheet. There is surprisingly little evidence about the explanation or instructions which Mr. Griffin gave Mr. Hawkins about the use of the spreadsheet. However, one can see from the document itself that it reflects the inventory which he had previously maintained in schedule form. For example, in common with his previous inventory the spreadsheet begins with the Kulab cargo and records the cargo sold under the first in-tank transfer as leaving the inventory on 25^th January, the date of the stock transfer certificate. I therefore infer that he simply transferred the existing data from his schedule onto the spreadsheet and continued to keep his records in the same way as he had done before. From early June the ITT contracts all contained a deemed stock transfer date for payment purposes and from about that time Mr. Hawkins treated oil as leaving the inventory on that date, i.e. the ITT date. The significance of this is a matter to which I shall return. However, it is important to note that this method of treating the ITT date as the date on which joint venture stock left the inventory came into existence at a time when both parties understood that in the absence of some special arrangements Glencore was only willing to sell oil to MTI on secured payment terms.
56. The December 1993 agreements were directed to a specific venture relating to cargoes of fuel oil which MTI had committed itself to buy from NIOC. The May 1994 storage agreement replaced the December 1993 storage agreement with effect from 18^th March 1994, but it was not in terms restricted to the 4-5 cargoes referred to in the original December agreement. For obvious reasons it did not contemplate segregated storage or redelivery of the specific oil discharged into storage and it was not of fixed duration. It seems clear, therefore, that at that stage the parties contemplated that further cargoes would be brought into storage under the agreement. No changes were made to any of the other elements of the December 1993 agreements, however, and I have little doubt that both Mr. Griffin and Mr. Heuzé proceeded on the basis that the existing arrangements would continue as before.
57. Between the middle of April and the end of September 1994 nearly forty parcels of 180cst and 280cst fuel oil were brought into Fujairah and disposed of under joint venture arrangements between Glencore and MTI. Some were sold on as cargoes to third party purchasers; others were sold by in-tank transfer to MTI. For the purposes of their record keeping and administration Mr. Hawkins and Mr. Griffin treated all this business as having been conducted under the same arrangements as those which had applied to the original cargoes stored on the Mount Athos. The evidence of Mr. Heuzé and Mr. Griffin and the documents as a whole leave me in no doubt that they saw this simply as a development of the original joint venture to be conducted on the same terms. However, it is fair to say that these were all Iranian straight run fuel oil cargoes and that to that extent they could naturally be seen as falling within the scope of the existing agreements. It is also true to say that separate storage and joint venture arrangements had been made for the two gasoil cargoes which were brought into Fujairah in January and June 1994.
58. One other matter needs to be mentioned at this point, namely, that throughout 1994 MTI was active as a supplier of bunkers in Fujairah, both to the local market and in the market for cargoes to the Far East, notably Singapore. Much of that bunker fuel was blended and substantial quantities of the Iranian fuel oil brought into Fujairah under the joint venture arrangements and sold by Glencore to MTI by in-tank transfer were used for the purposes of supplying bunkers. At the same time MTI had access to fuel oil and other products from suppliers other than Glencore. It is also perhaps interesting to note that the joint venture arrangements relating to the gasoil cargoes delivered on the Iter in January 1994 and the Macle in June 1994 were seen by the parties as a way of tapping the marine diesel bunker market in Fujairah. Mr. Kilakos said that MTI was entitled to draw and use that oil in small amounts before paying for it under ITT contracts for conveniently sized parcels, but that is contradicted by the documents which make it clear both that Glencore insisted on secured sales and that MTI was aware that Glencore's consent was required for the release of the cargo.

The Jaguar cargoes

59. The first cargoes of 380cst fuel oil brought into storage in Fujairah under joint venture arrangements between Glencore and MTI were those discharged from the Jaguar on 10^th and 15^th November 1994. MTI had previously obtained the oil needed to supply its bunkering business from a variety of sources and continued to do so. However, the development of their business relationship led Glencore and MTI to consider extending their joint operations into the purchase of bunker fuel for resale in the local bunker market. The primary use for 380cst fuel oil is as marine bunkers and it was common ground that at some point there had been an agreement to purchase cargoes of this kind with a view to resale in the local market. Mr. Griffin's recollection was that these two cargoes on the Jaguar were the first cargoes purchased specifically for that purpose. Mr. Heuzé could not recall whether that was so or not, but on balance I accept Mr. Griffin's account which seems to me to be supported by the pattern of cargoes coming into the joint venture around that time. Mr. Heuzé recalled that the initiative to extend the scope of the parties' business to include 380cst cargoes came from MTI. I think that is probably correct, since there was clearly an advantage for MTI, as there had been on previous occasions, to benefit from Glencore's financing of the oil. However, the proposal was also attractive to Glencore in that it offered an additional outlet for fuel oil together with an opportunity to obtain a share of the mark-up applied on the sale of bunkers into the retail market.
60. It had been an important part of the joint venture arrangements up to that time that the parties were obliged to consult each other about marketing cargoes held under the joint venture. Apart from anything else, that was essential to ensure that the same cargo was not sold twice. If parcels were sold to MTI for resale as part of its own bunkering business, the parties negotiated a price based on the prevailing market, but Glencore was not concerned in the disposal of the goods or the price at which they were sold on. Sales to MTI were treated in the same way as sales to any third party.
61. By the end of 1994 Mr. Heuzé was in frequent contact with Mr. Kilakos. Mr. Griffin's involvement was on the wane, but I accept his evidence that he was still in regular contact with Mr. Heuzé and that he remained well informed about the substance of discussions that took place between Mr. Heuzé and Mr. Kilakos. Mr. Crane put it to Mr. Heuzé in cross-examination that it had been agreed between himself, Mr. Kilakos and Mr. Griffin shortly before the purchase of the first of the two Jaguar cargoes that MTI should sell the cargo as bunkers and account to the joint venture in arrears. Mr. Heuzé firmly denied that and in the event none of MTI's own witnesses supported that case. It was the evidence of both Mr. Kilakos and Mrs. Gene that the new arrangement under which (as he said) MTI were authorised to sell bunkers drawn from joint venture stock and account for them in arrears had been introduced in about February 1995 following discussions he had with Mr. Heuzé and others from Glencore at the time of the International Petroleum Conference in London. Mr. Kilakos said nothing to suggest that an agreement of that kind had been made in the previous November. Mr. Griffin confirmed that Mr. Heuzé had never said anything to him to suggest that Glencore was willing to change its existing practice so as to allow MTI to dispose of oil brought into the joint venture before it had opened a letter of credit, nor was he aware from Mr. Kilakos of that having been said to him. It was not something that he would have expected Glencore to have agreed to in the absence of good commercial reasons and he did not suggest that any such reasons existed in the present case.
62. A good deal of attention was focused on a fax sent by Mr. Griffin to Mr. Heuzé on 21^st November in the following terms:

"Re: JV Sales – JV to Metro

Please note the following sales effected during the week of November 14-19. Effective ITT date November 21, 1994

12,000 mts      380cst      91.00

1,500 mts      380cst      90.25

13,000 mts      380cst      91.00

5,000      380cst      91.00

5,000      380cst      90.50

4,000      180cst      95.00

Sold today JV to Metro: - ITT date November 23, 1994

5,000 mts      180cst      94.00

5,000 mts      280      92.00

4,000 mts      380      90.75

Please confirm the above sales as usual by telex."

Mr. Crane suggested to Mr. Heuzé that, as he was well aware, the quantities mentioned in this message had been sold as bunkers from the joint venture inventory and that the quantities covered by the ITT contracts issued by Glencore tracked the amount of bunkers which Mr. Kilakos had told him had been sold each day. Mr. Heuzé, however, rejected that suggestion. He maintained that the message simply summarised daily sales of oil agreed between himself and Mr. Griffin during that week by telephone.

63. Mr. Griffin's account of this message has varied. In a deposition he gave in the United States in August 2000 he said that the document referred to sales which the joint venture had agreed to make to MTI during the course of the week in question and that the ITT dates mentioned in it were the dates on which the transfer of the cargo was to have taken place which established when payment became due. However, in a witness statement made for the purposes of these proceedings in February of this year he said that the expression "sales effected" referred to quantities of oil which had been sold by MTI as bunkers and that the sales by the joint venture were the agreed method by which MTI accounted to the joint venture for that oil. In cross-examination he initially adhered to that account, but later on he appeared to accept that the ITTs referred to sales by Glencore to MTI and he accepted that at no point had he himself made it plain to Mr. Heuzé that oil drawn from the joint venture stock had been used to supply bunkers to the market.
64. Since the products referred to in this message include 180cst and 280cst as well as 380cst fuel oil, any agreement of the kind suggested by Mr. Griffin would have had to have extended beyond the cargo on the Jaguar to include cargoes previously brought into Fujairah. However, in my view this and other similar messages sent by Mr. Griffin at around the same time speak for themselves. They plainly refer to sales by the joint venture to MTI, not to sales by MTI into the bunker market, although the quantities involved may well bear some relationship to the volume of bunkers sold by MTI on a day to day basis. However, since MTI had access to oil from sources other than the joint venture, it is impossible to know from this how closely the quantities it agreed to buy from the joint venture related to the quantities it had actually sold. The fact that parcels were purchased on a daily basis does not seem to me to point to an agreement that MTI was entitled to dispose of joint venture oil and account for it in arrears; it is equally consistent with a desire on the part of Mr. Kilakos to buy in stocks at a market-related price on a daily basis in order to keep pace with sales. At all events, two things emerge quite clearly from the evidence of Mr. Heuzé and Mr. Griffin. The first is that the profit-sharing principles of the original joint venture agreement established in December 1993 were treated as applying to the new cargoes brought into Fujairah during the middle and latter part of 1994. The second is that there is nothing to suggest that at this very early stage in the expansion of the business there had been any express agreement between the parties which would allow MTI to dispose of oil brought into the joint venture before it had become the subject of an ITT contract and a letter of credit had been opened in respect of the price.
65. I think Mr. Schaff was right in saying that any agreement to allow MTI to dispose of joint venture oil and to account to Glencore for it after the event would have involved a fundamental change in the nature of the parties' relationship. Not only would it have involved agreeing to trade on unsecured terms (a significant departure in itself); it would also have given MTI practical control of the stocks in the joint venture inventory. As Mr. Heuzé pointed out, at that time it was an essential aspect of the joint venture arrangements that the parties consulted each other over the sale of oil because they shared the responsibility for marketing it. Mr. Heuzé would undoubtedly have wanted to know how much oil was available for sale at any given time and the disposal of even relatively small parcels into the bunker market without prior warning could have posed significant problems. It is surprising, therefore, if MTI is correct, that there remains even now so much uncertainty over the time and manner in which a change of that kind occurred.
66. The fact that the form of the ITT contracts remained unchanged at this time is further evidence that the terms on which the parties were doing business remained unchanged. It is of course quite possible for businessmen to continue to use an established form of documentation even though it has ceased to reflect the true state of their relationship. Mr. Griffin suggested at one point that that is what happened in the present case in relation to accounting for products drawn to supply the bunker market, but I do not find that very persuasive. There are examples in the present case of the use of clearly inapposite documentation, but they are relatively isolated examples. If Glencore had consciously agreed to modify the basis on which oil was made available to MTI, I think it unlikely that it would have continued to use the existing form of ITT contract simply as an accounting procedure.
67. In my view all the evidence points clearly to the conclusion that there was no significant change in November 1994 in the basis on which the parties were doing business together. On the contrary, they continued to act on the basis that the existing storage agreement and joint venture agreement applied.

February 1995 - Discussions at the IP Conference

68. Between mid-November 1994 and early February 1995 a further ten cargoes of fuel oil were brought into store in Fujairah under the joint venture agreement and were disposed of under existing arrangements. It was Mr. Kilakos' evidence that discussions with Mr. Heuzé about the possibility of bringing bunker fuel into Fujairah which had begun many months before finally came to fruition in February 1995 when he was visiting London to attend the International Petroleum Conference. By then, he said, Mr. Heuzé had obtained the approval of his management to finance 300,000 tons of 380cst fuel oil a month to supply the bunker market. Although a number of people were involved in these talks on both sides, discussions about the essential principles were restricted to himself and Mr. Heuzé. In substance, he said that new terms were agreed under which MTI would buy the oil with financing provided by Glencore, MTI would charter the vessels needed to bring the oil to Fujairah, MTI would store the oil, blend it as it chose, and MTI and Glencore would together set sale prices for both bunker sales and cargo sales.
69. Mr. Heuzé agreed that he had met Mr. Kilakos in London in February 1995 and did not rule out the possibility of there having been discussions between himself and Mr. Kilakos relating to the expansion of the joint venture, although he doubted whether the whole matter had been raised and concluded all at once since their discussions normally developed in the course of numerous telephone calls over a longer period of time. He did accept, however, that at some point they had agreed to bring additional grades of oil into Fujairah with a view to providing products suitable for the bunker market. He disagreed sharply with Mr. Kilakos' suggestion that a new form of joint venture agreement had been put in place for that purpose. He said that the existing storage agreement and joint venture agreement remained in place.
70. One can see from the spreadsheet inventory kept by Mr. Hawkins that although most of the material brought into Fujairah for the joint venture during the following months was 180cst and 380cst fuel oil, other types of oil began to make an appearance as well. It seems likely, therefore, that the discussions between Mr. Heuzé and Mr. Kilakos at around the time of the IP Conference did lead to a gradual extension in the range of cargoes being handled. However, there remains a stark conflict of evidence as to the terms on which the joint venture was conducted thereafter.
71. In seeking to resolve this conflict it is necessary to consider both the documents which were generated in the course of joint venture dealings between Glencore and MTI and the account given by the witnesses, principally Mr. Kilakos, of the way in which the new system operated. As far as the documents are concerned, nothing changed. They continued to reflect transactions under which cargoes brought into Fujairah under the joint venture arrangements were either bought by Glencore from the prime supplier, or, more often, bought by MTI and sold to Glencore on back-to-back terms with Glencore providing the finance which enabled MTI to perform its purchase contract with the prime supplier. They continued to reflect arrangements under which cargoes were discharged either into the Metrotank or another vessel forming part of the storage facility as directed by MTI and held in common storage segregated only by grade before being sold to MTI on what had become the standard form of ITT contract. Each side continued to maintain an inventory in the same manner as before and continued to reconcile the financial outcome in the same way as before. On the face of the documents, nothing changed.
72. According to Mr. Kilakos, however, the terms on which the joint venture was conducted changed quite substantially. From about February 1995 a major focus of the joint venture was the supply of bunkers into the local market (although MTI continued to buy oil for its own account from the joint venture) and a method of working was introduced to deal with that which called for discussions between himself and Mr. Heuzé twice a day. Each evening he and Mr. Heuzé would discuss the market and agree upon a selling price for the following day and the maximum quantity to be sold on behalf of the joint venture. The next day Mr. Kilakos would give appropriate instructions to the traders and he and Mr. Heuzé would review the position at around noon. Each day MTI would sell a quantity of bunkers and subsequently by means of an ITT contract would credit the joint venture with the price at which they had been sold, less MTI's expenses which would range from $1.25 to $2.5 per ton. They did not discuss the mechanics of blending because that was something within Mr. Kilakos' special expertise, although Mr. Heuzé knew that bunkers were being blended because the nature of the oil being brought into Fujairah for the joint venture made that a necessity. Because bunkers are sold in fairly small quantities daily sales were often grouped together and ITT contracts issued in rounded off amounts. Moreover, this method of doing business was extended to those cases in which MTI bought oil for its own account from the joint venture so that in such cases MTI was permitted to dispose of the oil straightaway if it had an immediate need to satisfy a purchaser.
73. Although during this time MTI was, according to Mr. Kilakos, selling bunkers on behalf of the joint venture, it was also selling bunkers on its own behalf and purchasing fuel oil and other products from Glencore and other suppliers in order to do so. Mr. Kilakos estimated that during 1995-1996 transactions with Glencore represented only about 40% of MTI's business. One might have thought that a business which was being run in the manner he suggested would need to record transactions made on its own behalf separately from those made on behalf of the joint venture, but in fact there is nothing in MTI's records to distinguish transactions made for the joint venture from transactions made for its own account. Inventories were kept of oil available to MTI (about which I shall have to say more later), but these drew no distinction between the sources from which such oil had been obtained. Similarly, no records were kept of bunker sales which would enable those made on behalf of the joint venture to be distinguished from those made by MTI for its own account. On the other hand, MTI's records accord perfectly well with a system under which oil was purchased for its own use from various sources including the joint venture at prices fixed by negotiation based on market conditions. Mrs. Gene would often be informed that Mr. Kilakos had agreed to buy a quantity of oil from the joint venture and would confirm with him the quantity and price so that she could confirm the accuracy of the ITT contract when she received it. I have no doubt that Mr. Kilakos sometimes told her that the transaction reflected quantities of oil he had sold in the bunker market, but, as she admitted, she was not told whether the oil he had agreed to buy was related to any particular business. She had no way of knowing, therefore, which parcels were bought for which particular purposes.
74. Similarly, the idea that daily sales made at different prices should be grouped together and accounted for by ITT contracts in "rounded off amounts" seems to me to be both unnecessary and unbusinesslike and thus a poor explanation for the form in which the documents are found. There is no obvious reason why sales of whatever quantity could not be accounted for in a simple and straightforward way, whereas "rounding off" was bound to introduce inaccuracies. However, if the parties had simply intended that MTI should account to the joint venture for oil sold on its behalf, there is no obvious reason why a different and much simpler form of documentation should not have been devised. I can understand that for record keeping purposes Glencore might have wanted a document that purported to pass title in the oil to MTI, but it is difficult to understand why it should have continued to make use of the existing form of ITT contract which was so plainly inappropriate. On the other hand, if MTI was in fact purchasing oil for its own account, the continued use of the established form of ITT contract makes perfectly good sense. It is also understandable that MTI should purchase "rounded off" quantities broadly reflecting its sales over a period of some days rather than many small parcels reflecting day to day sales.
75. Although it did not form part of MTI's pleaded case, Mr. Kilakos said in evidence that he made a new storage agreement, or "throughput agreement" as he preferred to call it, with Mr. Heuzé during their discussions in London in February 1995. Instead of charging Glencore $1.25 per ton for the first 20 days and $0.05 per ton per day thereafter, MTI agreed to accept $1 per ton as a flat fee on all cargoes brought into the floating storage. However, here too the documents fail to show any sign of a change in the parties' arrangements. In fact a reduced storage fee of $1 per ton for the first 20 days and $0.05 per ton per day thereafter had been introduced in July 1994 with effect from 16^th May. Because oil was held in commingled storage any assessment of the period for which a particular cargo remained in storage could only be calculated on a notional basis. Calculations carried out in MTI's office at an early stage on a 'first in, first out' basis indicated that oil rarely remained in storage for more than 20 days. The calculation was difficult to perform and Mrs. Gene considered it a waste of time to work out the precise storage period in each case simply to enable MTI to charge Glencore a few cents per ton on any quantities which remained in store for more than 20 days. The exercise was therefore not repeated. From the autumn of 1994, therefore, Glencore was simply charged $1 per ton on all cargoes discharged into the floating storage. A time did come in April 1996 when MTI described this fee of $1 per ton as a "handling fee" in the context of a dispute over whether it should be paid on all cargoes arriving at Fujairah, whether they were discharged into storage or not. However, that does not support the conclusion that the storage arrangements were modified in February 1995 as Mr. Kilakos suggested. On the contrary, had there been an agreement for a flat rate throughput fee in February 1995, that dispute would never have arisen. A note made by Mr. Heuzé in January 1995 might suggest that there had been some discussion about the $1 fee at that time, but Mr. Heuzé himself was not asked to comment on it and when it was shown to Mr. Kilakos he said that it did not remind him of any conversation. In these circumstances I am unable to derive an assistance from it.

Conclusion

76. In these circumstances I am satisfied that the extension of the joint venture represented by the decision to purchase new grades of oil suitable for resale in the bunker market, either in their original form or after blending, was not seen by Mr. Heuzé or Mr. Kilakos as a radical departure from their existing business but as a natural extension of it. During 1994 they had continued to do business on the basis of the December 1993 and May 1994 agreements, discarding those terms which were no longer apposite but retaining the fundamental principles. There was no difficulty in applying those principles to the purchase and sale of new grades of oil and they continued to work together on the understanding that they would continue to apply. There was no need in November 1994 or February 1995 to agree new terms, and none were agreed.

The course of dealing, February 1995 - June 1996

77. Despite the absence of any explicit agreement in February 1995 on new terms to apply to the disposal of oil destined for the bunker market, both Mr. Crane and Mr. Cooke submitted that an examination of the course of dealing between the parties over the ensuing months shows that they did in fact depart quite significantly from the terms originally agreed, so much so that they can be seen to have agreed to act on terms very different from those which originally governed their relationship. It is necessary, therefore, to turn next to the conduct of the parties between November 1994 and June 1996 to see what inferences can properly be drawn from it.
78. Mr. Crane drew attention to the five parcels of 380cst fuel oil sold to MTI under the ITT contract dated 23^rd November 1994 with an ITT date of 21^st November. These, of course, are the parcels of oil referred to in Mr. Griffin's message of 21^st November 1994 which I quoted earlier. He submitted that this and later messages of a similar kind reflected what became an established practice under which MTI sold as bunkers oil drawn from the joint venture inventory for which it subsequently accounted through the mechanism of ITT contracts.
79. I have already explained why I am unable to accept that there was any explicit agreement between Mr. Heuzé and Mr. Kilakos that MTI should dispose of oil from the joint venture inventory in that way and why I am unable to accept that the parties conducted themselves on that understanding during the latter part of 1994 and the early part of 1995. However, Mr. Crane pointed to a number of aspects which he submitted pointed to the conclusion that the parties had agreed to act in that way. I have already considered some of these, but one which I have not yet touched on is the phenomenon of what was called "negative inventory".

(i) Negative inventory

80. The computer generated spreadsheets kept by Glencore and MTI recorded the movements of oil into and out of the joint venture grade by grade. They did not include a running record of the quantity remaining in store as a result of these transactions, but they did contain a record of the quantity currently remaining in store which was capable of being automatically updated from time to time. The spreadsheets, in common with other documents recording the movement of oil, treated oil as entering the joint venture inventory on the date of completion of discharge at Fujairah and as leaving the joint venture inventory on the date of the outward bill of lading in the case of cargo sales and the ITT date in the case of in-tank transfers to MTI. An analysis of the movements recorded in Mr. Hawkins' spreadsheets shows that on various occasions the joint venture inventory was negative, that is, more oil was recorded as having been withdrawn from the joint venture stock than had been recorded as available. In some cases a single transaction accounted for the stock's being overdrawn; in other cases withdrawals are recorded as having been made at a time when stocks were already overdrawn. This phenomenon of being overdrawn became known during the trial as "negative inventory".
81. The parties were well aware that on occasions more oil was required to meet obligations entered into for the account of the joint venture than was available for that purpose. However, MTI had its own stocks of oil in storage which could be drawn upon to make up the shortfall. On some occasions, therefore, in order to complete a cargo oil was drawn from MTI's own stocks which were replenished by a delivery for the account of the joint venture a few days later. It must be born in mind that since oil of the same grade was commingled in storage, identification of the amount held for different interests was essentially a matter of accounting. One can find instances of this occurring from as early as the summer of 1994.
82. There were other occasions, however, on which the transaction which took the joint venture inventory into the negative was an in-tank transfer to MTI, and in some cases one can find several in-tank transfers recorded in the spreadsheets at a time when the inventory was already in the negative. However, it is clear that the parties themselves were aware of the position because from time to time steps were taken to bring the inventory back into balance, usually by blending a parcel whose constituents were drawn from other grades held for the account of the joint venture. It will be necessary to consider at a later stage the implications of treating oil as leaving the joint venture inventory at the ITT date, but accepting for the moment that the movements recorded in the spreadsheet reflect physical movements of stock, what does it show? Mr. Crane submitted that it shows MTI supplying to third parties in the form of bunkers oil which by agreement between the parties was attributed to the joint venture even though it had no remaining stock. The fact that MTI paid Glencore through the mechanism of the ITT contracts in respect of an apparent transfer of non-existent stock demonstrates, he submitted, that MTI was in fact accounting to the joint venture in respect of oil already drawn from what was agreed to be the joint venture stock.
83. The fact that by the latter part of 1994 MTI was holding all fuel oil at Fujairah in commingled storage segregated only by grade means that any inventory of the kind kept by Mr. Hawkins is essentially in the nature of an accounting document. Provided sufficient physical stocks are held to cover the interests of all the contributors it can be said that an inventory of this kind records the physical stock owned by the depositor, although what it actually records is the quantity of oil which the depositor is entitled to withdraw from the bulk. If the depositor wishes to withdraw more oil than is currently available to him, he must obtain the agreement of another depositor to draw from the oil available to him. If some of the oil belongs to the operator of the store, as it did in this case, drawing from his stock will create a negative balance as between the depositor and the operator. That is clearly what happened from time to time in the present case, as Mr. Crane conceded, when Glencore's inventory went into the negative in order to complete a cargo being shipped out of Fujairah for the account of the joint venture. In such cases MTI simply drew the amount of oil required from its own stocks and both parties treated the joint venture as "overdrawn" to that extent. The overdraft was treated as automatically repaid when more oil of the same grade was brought into store for the account of the joint venture. There does not appear to have been any specific discussion about how to handle the situation, much less any express agreement to deal with it in that way; the parties seem to have treated it as the natural way in which to operate.
84. There is no difference in principle between the situation in which MTI lends oil to the joint venture to complete a cargo and that in which it lends oil to the joint venture to dispose of in any other way, including for sale to itself by in-tank transfer. Of course it can be said that it is anomalous for the joint venture to sell oil to MTI which both parties know that it does not have and for MTI to pay for it under a letter of credit, but it is more understandable when one bears in mind that Glencore was financing the joint venture for the benefit of both parties and that since in all cases the deficit was made good at regular intervals the financial effect of these transactions was the same as if the necessary stock had been available at the appropriate time. I am therefore unable to accept Mr. Crane's submission that the creation or enlargement of negative inventory by in-tank transfers necessarily points to the conclusion that the parties had agreed that MTI should dispose of joint venture stock for which it would account in arrears.

(ii) Cargo sales – the "Jag Laadki", the "Hellespont Faith" and the "Ouranos"

85. During the course of JV1 there were a number of occasions on which the established form of ITT documentation was used to cover transactions between Glencore and MTI relating to cargoes delivered out of the floating storage facility. Mr. Crane submitted that they demonstrate that Glencore was quite willing to use that form of documentation in relation to oil which was clearly not in tank at the time the contract was made and to issue stock transfer certificates even though it was well aware that the oil had long gone. They also show, he submitted, that the cargoes in question were treated as having been blended using joint venture stock before any security in the form of a letter of credit had been put in place.
86. The first example concerned a cargo of about 125,000 tons of 380cst loaded on the vessel Jag Laadki at Fujairah in April 1995. The cargo was loaded under bills of lading dated 15^th April pursuant to four contracts of sale between MTI and various purchasers. On about 25^th April, by which time the vessel was only one day out of Singapore, it was agreed that the joint venture should participate in the contracts to the extent of 50% and Mr. Hawkins recorded in his spreadsheet an in-tank transfer by the joint venture to MTI of 62,361.980 tons of oil (half the total cargo) on 15^th April. On 30^th May Glencore issued an ITT contract in the usual terms describing the oil as "stored in the vessel Metrotank off Fujairah" and providing for payment by letter of credit of a price calculated to reflect the contract value of the oil f.o.b. Fujairah. In the event MTI paid for the parcel in cash on 2^nd June on receipt of an invoice from Glencore and no letter of credit was opened. On the same date Glencore issued a stock transfer certificate in favour of MTI.
87. Mr. Heuzé took issue with the suggestion that the parties had agreed after the event to treat half the cargo as having been drawn from the joint venture stock. He said that he had simply agreed to sell MTI a quantity equivalent to half the cargo which it had sold into Singapore on the Jag Laadki and that the sale was secured inasmuch as the oil had only been released to MTI by the issue of a stock transfer certificate against payment. I do not find that a very persuasive explanation. The telex which was sent confirming the arrangement speaks of the joint venture's "participating in" these sales and Mr. Hawkins entered the covering in-tank transfer in his spreadsheet as taking effect on 15^th April. That is a clear indication that Glencore was treating the transfer as having taken place on that date as if it were the ITT date. I think Mr. Crane was right, therefore, in saying that on this occasion Glencore was unsecured for the price of the oil between 25^th April and 2^nd June.
88. Another transaction of the same kind occurred in April 1996 when MTI agreed to sell 80,000 tons of 380cst fuel oil to Kuo Oil. A cargo of 83,510 tons was shipped on the Hellespont Faith by MTI on 17^th April under a bill of lading consigned to its own order. On 22^nd April Mr. Heuzé agreed to sell MTI approximately 50,000 tons of oil from the joint venture stock at a price based on the average of Platts Singapore quotations for various dates around the vessel's notice of readiness at Singapore. That agreement was confirmed by an ITT contract sent on 26^th April 1996 which described the product sold as "fuel oil 380cst stored on vessel m/t Metrotank off Fujairah loaded onto mv Hellespont Faith b/l date 17/4/96". The vessel arrived at Singapore on 29^th April and completed discharge on 30^th April. Payment was made under a letter of credit opened in favour of Glencore on 7^th May and a stock transfer certificate was issued on 21^st May. The oil covered by this contract appears in Mr. Hawkins' spreadsheet first as a blend of other materials producing an addition of 50,000 tons to the joint venture stock of 380cst fuel oil and secondly as an in-tank transfer of that quantity to MTI on 17^th April.
89. Mr. Heuzé insisted that there was no physical link between the oil sold by MTI to Kuo and the oil sold by the joint venture to MTI. He pointed to the pricing formula in the ITT contract which was based on the Singapore market price less $5.50 representing the freight differential between Fujairah and Singapore. That showed, he said, that Glencore was selling f.o.b. Fujairah, not c. & f. Singapore as MTI had done and that MTI was purchasing oil after the event to replenish its own stock.
90. It was no part of MTI's case that in April 1996 it was entitled to draw oil from the joint venture stock at will in order to make cargo sales to third parties. Under JV1 the marketing of cargoes was a matter for agreement between the parties. MTI's records show that at the time the Hellespont Faith was loaded on 17^th April it had over 450,000 tons of fuel oil in stock, so I do not think that there can be much doubt that the cargo was originally drawn from its own inventory. I think Mr. Heuzé was right, therefore, in saying that there was no physical link between the two sales in the sense that MTI had not drawn on joint venture stock in order to load the vessel. The contract under which Mr. Heuzé agreed to sell 50,000 tons of fuel oil to MTI was, of course, directly related to the sale to Kuo and the pricing mechanism meant that the joint venture participated in the market risk involved. However, it remained a sale of oil situated in Fujairah. The only question of importance for present purposes is whether Glencore intended to transfer title in that oil to MTI immediately or only at a later date.
91. The documents relating to this transaction all support the conclusion that the parties intended to treat this parcel of 50,000 tons as having been part of the cargo originally loaded on board the Hellespont Faith. Mr. Heuzé's deal ticket does not mention the vessel specifically, but it does refer to the expectation that the bill of lading would become available on 18^th April. The ITT contract, on the other hand, does specifically refer to the oil as having been loaded onto the vessel and both Mr. Hawkins' spreadsheet and MTI's oil records contain entries which treat that part of the cargo as having come from the joint venture stock. MTI later sent Glencore a note of the blendstock which was said to have been used to make up the parcel shipped on the Hellespont Faith which in turn provides the basis for the blend entry in Mr. Hawkins' spreadsheet which I mentioned a little earlier. I shall return to the other implications of that later on, but in the present context it supports the conclusion that both parties agreed to treat the oil as if it had been part of the original cargo. I think Glencore must have realised that MTI would be the original holder of the bill of lading and that property in the goods would pass to Kuo with effect from the point of loading once the bill of lading was taken up under a letter of credit. This is the normal way in which transactions of this kind are structured and in any event MTI sent Glencore a copy of its contract with Kuo shortly before the ITT contract was issued. In these circumstances I find it difficult to believe that Glencore did not intend title to the oil to pass to MTI at the latest when the bill of lading was transferred under the sale to Kuo. To that extent the provisions in the ITT contract dealing with the transfer of title are inapposite and Glencore was unsecured in respect of the value of this parcel of oil for at least the period from 30^th April until the letter of credit was opened on 7^th May.
92. A third example of this kind of agreement occurred a few days later in relation to a cargo of 80,000 tons of 180cst fuel oil loaded on the vessel Ouranos on 20^th April also pursuant to a contract of sale by MTI to Kuo Oil. On 26^th April Glencore issued an ITT contract to MTI in respect of 50,000 tons of oil "stored on vessel m/t Metrotank off Fujairah loaded onto mt Ouranos b/l date 20/4/96". The documents in this case are all worded in very similar terms to those relating to the Hellespont Faith parcel. The provisions relating to the calculation of the price were substantially the same as those agreed in relation to the previous parcel so the commercial effect was essentially the same. In this case the vessel arrived at Singapore on 2^nd May and completed discharge on 4^th May. Payment was made under a letter of credit opened in favour of Glencore on 7^th May and a stock transfer certificate was issued on 21^st May. As in the previous case, the oil covered by this contract appears in Mr. Hawkins' spreadsheet first as a blend of other materials producing an addition of 50,000 tons to the joint venture stock of 180cst fuel oil and secondly as an in-tank transfer of that quantity to MTI on 20^th April. A blending advice identifying the components used to produce the parcel of 180cst oil loaded on the Ouranos was sent to Mr. Hawkins on 2^nd May. For the same reasons as in the case of the Hellespont Faith parcel I am satisfied that the parties intended that property in the oil should pass to MTI at the latest when the bill of lading was transferred to the buyer and that as a result Glencore was unsecured in respect of the price of this parcel for a period of a week or more.
93. What significance is to be attached to these transactions? Mr. Crane submitted that they are examples of Glencore's willingness during the period prior to June 1996 to sell oil to MTI otherwise than on secured terms and that as such they undermine the contention that Glencore would never have contemplated allowing MTI to dispose of oil from the joint venture stock before some kind of security, normally a letter of credit, had been put in place. This is a fair point, but it should not be over-stated. These were but three transactions out of some hundreds which the parties entered into between February 1995 and June 1996. They are not typical of the way in which the parties did business and it is telling in my view that none of MTI's witnesses said in their statements that they had regarded them as significant for that reason. It has never been disputed that by 1996 a substantial measure of trust had developed between the parties, so it is not surprising that from time to time Glencore should have been willing to act in that way, even if secured transactions remained the general rule.

(iii) Cargo sales by MTI on behalf of the joint venture

94. From time to time MTI sold cargoes of oil to third parties on behalf of the joint venture. This involved the sale and delivery of oil which both parties understood to be held in store to the order of Glencore. Mr. Crane drew attention to eleven such transactions, six involving the vessel Aitolikos, four involving the vessel Fay and one involving the vessel Honam Saphire. These transactions all followed a broadly similar pattern. MTI entered into a contract to sell oil c. & f. Singapore on terms that property in the goods was to pass on loading. Payment was to be made by letter of credit, or, in the case of sales to one of the major oil companies, by telex transfer. Shortly afterwards Glencore entered into a back-to-back contract with MTI for the sale of the same quantity and quality of oil except that payment was to be made by letter of credit or other acceptable security. In practice MTI did not open a letter of credit but assigned the benefit of the relevant payment obligation to Glencore and instructed the bank to which the price was due to be paid to pay the amount due to Glencore.
95. It was inherent in this way of doing business that Glencore did not obtain security for the payment of the price payable by MTI until an assignment had been made of the relevant payment obligation. In most cases the cargo was shipped within a few days of the date of the contract between Glencore and MTI, but in three cases the cargo had already been shipped by the time the contract was made and in one case it had even been delivered to the purchaser. What is perhaps more significant for present purposes is that security in the form of an assignment of the proceeds of sale or the sum due under the buyer's letter of credit was not in place until some days after the cargo had been shipped and in a few cases had not been put in place until after the oil had been delivered to the buyer.
96. The transactions relating to the cargo shipped on the Honam Saphire followed this general pattern except that three parcels were loaded under three separate contracts between MTI and Sanko Oil. The contracts between them for the sale of the first two parcels were made in the latter part of December 1995. Mr. Heuzé's deal tickets relating to the corresponding sales by Glencore to MTI are both dated 7^th December, although the contract confirmations were not sent until 22^nd December. The contract between MTI and Sanko relating to the third parcel must have been made just before or just after the vessel completed loading at Fujairah on 16^th January 1996. MTI sent a copy of the contract confirmation to Glencore on 17^th January but Mr. Heuzé's deal ticket relating to the corresponding transaction between Glencore and MTI is dated 15^th January. As is usual, each of these contracts provided for title to pass on loading. Sanko's letter of credit relating to the first two parcels was in place before the vessel left Fujairah, although it was not assigned to Glencore until 19^th January, by which time the vessel had already sailed. The letter of credit covering the third parcel was not opened until 26^th January and was not assigned to Glencore until 29^th January when the vessel was already at Singapore.
97. In the case of all these transactions MTI was named as shipper in the bill of lading and had control over the oil once it was on board the vessel. Glencore had no security for the payment of the price until it received an assignment of the price due from Sanko to MTI or an assignment of the benefit of the letter of credit. Mr. Crane made a number of criticisms of the way in which Mr. Heuzé and Mr. Hawkins had dealt with these shipments on the Honam Saphire. In their witness statements they had each said that oil was not released from storage until MTI had provided security in the form of an assignment of the proceeds of sale. Mr. Crane submitted that they had given a one-sided account of the transactions which could only be maintained by exhibiting a selection of the documents. He also accused them of demonstrating what he described as a dogged determination to maintain an agreed line in the face of irrefutable documentary evidence to the contrary such as to undermine the value of their evidence as a whole.
98. In the case of the Honam Saphire I think in retrospect there are grounds for criticising the choice of material which was exhibited to Mr. Heuzé's witness statement, but I am less certain how much real blame can be attached to him, or indeed anyone else, for that. A huge amount of paperwork was generated in the course of the joint venture and the task of sifting and marshalling it for the purposes of the trial must have been daunting for all parties. The transactions on which I am currently focusing represent only a small fraction of the total and the fact that one can see at this stage after the material has all been carefully scrutinised that omissions have been made does not inevitably lead to the conclusion that the original selection was made in bad faith. I am not persuaded that that is what happened in this case. It is fair to say that both Mr. Heuzé and Mr. Hawkins displayed to differing degrees a tendency to argue Glencore's case as they saw it, but the same criticism can be made of Mr. Kilakos and Mrs. Gene. That is unfortunate, no doubt, but understandable, particularly in the case of those who, like Mr. Heuzé and Mr. Kilakos, were not only deeply involved at the time but have since spent innumerable hours labouring over the documents for the purposes of these proceedings.
99. The main explanation that Mr. Heuzé gave for the state of affairs illustrated by these transactions is that from a commercial point of view the nature of the risk involved was essentially different from that which existed when Glencore sold to MTI for its own consumption. The real credit risk, as he saw it, was that of the end buyer, not MTI. In some cases the buyer's letter of credit was in place before the cargo left Fujairah, so the only risk which Glencore bore was that of a failure by MTI to provide the assignment. Mr. Heuzé thought that Glencore could exercise a degree of practical control over the cargo in its capacity as charterer of the vessel. Whether he was right about that or not does not really matter if, as I accept was the case, it affected his view of the risk involved in the transaction. In other cases where a letter of credit was not already in place MTI as holder of the bill of lading controlled the cargo and it had a strong incentive to ensure that the buyer opened its letter of credit as soon as possible.
100. Mr. Crane submitted that these transactions provide further examples of Glencore's willingness, contrary to what was said by Mr. Heuzé and Mr. Hawkins, to deal with MTI on an unsecured basis. In one sense that may be true, though I can see why Glencore regarded these transactions as commercially rather different from those under which it simply sold goods to MTI, and I can also understand that for different reasons Glencore was willing from time to time to take a commercial risk if the circumstances warranted it, for example where the end buyer was known and trusted. It is a far cry from that, however, to say that Glencore was willing to accept as part of its routine trading relationship with MTI sales of oil on open account terms, much less that Glencore was willing to allow MTI to draw on the joint venture stock at its discretion for the purposes of blending and supplying bunkers and to account for what it had drawn only after the bunkers had been sold.

(iv) Blending

101. A great deal of the oil brought to Fujairah under JV1 was heavy oil for which there was no generally available market but which was suitable for blending into products for sale in the commercial market as fuel oil. Consistently with its allegation that MTI was not entitled to dispose of oil bought by in-tank transfer before a letter of credit had been opened in respect of the purchase price, Glencore contended that any blending of oil drawn from the joint venture stock without its express approval before a letter of credit had been opened was, and was known to be, a wrongful misuse of its property. MTI, on the other hand, maintained that it had a general right to blend oil from the joint venture stock whenever it considered it in the interests of the joint venture to do so. I turn next, therefore, to consider the circumstances in which blending took place and the extent to which it was known to and acquiesced in by Glencore.
102. It is necessary to begin by clarifying the terms used in this context. For present purposes the primary distinction to be drawn is between the mixing in common storage of different parcels of oil of substantially the same grade and the deliberate admixture of two or more parcels of oil of different grades to produce a single parcel of oil of a grade substantially different from any of its original constituents. The former is usually described as "commingling", the latter as "blending" and in the course of the trial these two terms were used in the senses just described.
103. Both parties understood that under the joint venture arrangements different cargoes of oil of the same grade would be commingled in common storage. They were also well aware that once discharged into common storage the individual identity and characteristics of any particular cargo would be irretrievably lost, but within grades those individual characteristics were not of any commercial significance and both parties were therefore content to treat one parcel of oil of a given grade as equivalent to another. Blending, whether into a finished product or as a step on the way to producing a finished product, is clearly a different matter altogether, however, because the commercial characteristics of each contributing parcel are irretrievably lost in the blend and a new product is created which is commercially different from all of its various constituents.
104. In deciding whether the characteristics of any given parcel of blended product are different from those of its constituents it is necessary to have regard principally to the commercial context and the intention of the parties. In the present case the individual grades of cargo were established by agreement between Glencore and MTI and were reflected in their respective inventory spreadsheets. Even though the components of a blend are not physically destroyed in the blending process, the use of oil as a blendstock must in my view be characterised as a form of disposal because it involves the irretrievable loss of the separate identity and characteristics of the original components. A similar loss of characteristics or even of identity may, of course, occur inadvertently, as for example, when there is leakage of oil from one tank into another. This would normally be regarded as an example of contamination (or in an extreme case destruction) rather than blending and raises questions of a different kind. Such questions do not arise in the present case, however, and therefore do not call for consideration.
105. Mr. Kilakos was known to have a particular expertise in blending and from an early stage a number of cargoes were blended for shipment out of Fujairah for the account of the joint venture. Mr. Crane submitted that Glencore was willing to give MTI a great deal of discretion to blend oil drawn from the joint venture stock whenever Mr. Kilakos thought it commercially sensible in order to produce products for sale on behalf of the joint venture. Indeed, at one point he went so far as to submit that routine blending must have been implicit in the joint venture arrangement unless expressly prohibited.
106. That blending was a frequent occurrence during 1995 and the first half of 1996 is evident from the documents, in particular the inventories and reconciliation sheets produced by the parties for the purpose of accounting to each other for joint venture profits and losses. Over two thirds of the cargoes that left Fujairah for the account of the joint venture during that period had been blended and a blending fee was charged by MTI to the joint venture. Mr. Crane drew attention to the fact that there are no documents specifically authorising any of that blending and submitted that the evidence as a whole shows that Glencore simply allowed MTI to carry out whatever blending operations it thought necessary. In my view, however, no such inference can be drawn. During this period all sales had to be the subject of discussion and agreement between Mr. Heuzé and Mr. Kilakos. Mr. Heuzé's agreement to the sale of a cargo to a third party necessarily carried with it Glencore's authority to deliver that cargo out of the joint venture stock. In some cases where blending was necessary to produce the required product Mr. Heuzé and Mr. Kilakos may have discussed the components required for the blend, in others they may not, but in either case the inference must be that Glencore authorised MTI to make up the blend from any appropriate components held in the joint venture stocks. In my view there is nothing in these transactions to support the conclusion that MTI enjoyed a more general discretion to blend.
107. Apart from blending to produce cargoes for sale to third parties, there are recorded in the spreadsheets and other documents several instances of the blending of oil for sale by Glencore to MTI by in-tank transfer. All these so-called "ITT blends" were documented in a broadly similar way and it is therefore sufficient to describe them by reference to a few striking examples. On 28^th May 1996 Mrs. Gene sent a telex to Mr. Hawkins asking him to amend his inventory to reflect 11 in-tank transfers totalling 110,000 tons of 380cst fuel oil agreed between 8^th and 23^rd May with ITT dates ranging from 8^th to 27^th May. In the same telex she identified the components which had been used to produce a total of 110,000 tons of 380cst oil in four separately blended parcels. Mr. Hawkins duly entered these 11 in-tank transfers in his spreadsheet by reference to their respective ITT dates (even though no letter of credit had been opened in respect of any of these quantities by that date) and entered into the spreadsheet, undated but immediately after an entry dated 27^th May, four "adjustments for blends" showing withdrawals from the relevant component grades and additions to the stock of 380cst fuel oil.
108. Similar adjustments to inventory were carried out in June and October 1996. In each case Mr. Hawkins entered the adjustments in his spreadsheet on or shortly after receiving the telex advice from Mrs. Gene, but entered the individual in-tank transfers on their respective ITT dates.
109. The language of the telexes sent in May and June 1996 was ambiguous: they simply asked Mr. Hawkins to amend his inventory to "reflect" joint venture sales to MTI, thereby leaving open the possibility that these were what Mr. Heuzé and Mr. Hawkins described as "notional" blends, that is, merely transfers of the components that would be needed to produce the required quantity of 380cst fuel oil. The telex of 7^th October is more explicit, however, in that it referred to "Quantities/grades used for blending 380cst on 07/10/96" (my emphasis). This particular message referred to a total of 160,000 tons of 380cst fuel oil sold to MTI under 16 separate in-tank transfers with ITT dates ranging from 14^th September to 7^th October. Mr. Hawkins said that he did not notice the word "used" in this and other similar telexes.
110. The importance of this evidence lies in the light it throws on Glencore's understanding of how MTI was making use of joint venture stocks for blending. Mr. Hawkins insisted that MTI was not entitled to take any stock for blending unless and until a stock transfer certificate had been issued in respect of it and that he would have protested to Mrs. Gene if he had thought that MTI was not observing that requirement. He maintained that only after a stock transfer certificate had been issued was MTI entitled to blend the parcel in question from the components identified to Glencore, or if it so chose, to take the components in their original form for its own use.
111. The manner in which Mr. Hawkins recorded the individual transfers in his spreadsheet by reference to their ITT dates is consistent with his general practice, but the entry of the blended parcel as an addition to stock on the date when he received the advice from Mrs. Gene is only consistent with his treating the whole of the blended parcel as having been produced by that time. However, at the time these entries were made letters of credit had been opened and stock transfer certificates had been issued for few, if any, of the blended parcels, so that if he was right, no blending should in fact have occurred by that time. One can rationalise this method of book-keeping by saying that transfers out of inventory of parcels of blended product need to be matched by corresponding transfers into the inventory, but this makes the whole exercise very artificial. Since the adjustments were recorded several days after the dates of the earliest in-tank transfers, there is no obvious reason why they should not have been recorded as and when individual parcels were blended or following the blending of the whole parcel.
112. However, all this has to be viewed against the background of commercial operations at Fujairah. Mr. Kilakos said that quantities of the kind referred to in these telexes would normally have been blended as single parcels over a period of anything up to two days depending on where the components were stored and what other work had to be undertaken. What seems clear is that a series of in-tank transfers of the kind covered by the telex of 7^th October would not have been blended individually. Whether that was something of which Mr. Hawkins was aware from other sources does not matter because the information given in the telex itself indicates that a single blending operation was involved. Moreover, the way in which the telex is worded naturally indicates that a blending operation had already been carried out to cover the contracts to which it refers with a surplus of 10,000 tons not covered by any existing contract. In all the circumstances I find it impossible to accept that in this case Mr. Hawkins did not realise that MTI had already carried out the blend to which the telex refers.
113. In his supplementary statement Mr. Heuzé sought to explain these transactions as being sales of components rather than sales of blended product. He said that because the components had to be priced by reference to the market price of the generic grades for which a price had in turn to be negotiated by reference to the Singapore market price, it was simpler to agree prices for the generic grade represented by the blend than for the individual components. He said that in each case he agreed with Mr. Kilakos what components would be used to produce the required blend and was told by Mr. Kilakos what quantities would be required for that purpose. In cross-examination he first indicated that he did not really know or care whether the blend had actually taken place because property in the product would not pass to MTI until a stock transfer certificate had been issued. Until that time MTI was bound to hold it to Glencore's order. (Whether he was right or wrong about that, it is difficult to reconcile with Glencore's case that MTI was not entitled to make use of any components for blending until a stock transfer certificate had been issued in respect of them.) Later in his evidence, however, he said that he was convinced the blends referred to in these telexes had not in fact taken place and were purely notional.
114. There may have been cases in which blends were purely notional - the stock transfer carried out in March 1996 in exchange for the cargo of A960 carried on the Kingfisher may be one – but I find this explanation difficult to reconcile with the documents to which I have just referred. In the first place, if MTI wanted to purchase certain quantities of components one would expect that to determine the quantity of the notional blend, rather than the other way round. More importantly, however, the pattern of ITT blending does not reflect a desire on the part of MTI to purchase components rather than finished products. I can well imagine Mr. Heuzé and Mr. Kilakos making agreements at intervals of a few days for the sale of parcels of 10,000 tons of fuel oil which were then produced by a single blending operation, following which MTI notified Glencore of the precise quantities of components used for the purpose. I find it much more difficult to imagine Mr. Heuzé and Mr. Kilakos making agreements at intervals of a few days for the sale of components which subsequently led to successive ITT contracts for the sale of parcels of 10,000 tons of fuel oil.
115. The difficulties in understanding the way in which the parties treated ITT blends largely disappear once one accepts that they treated them in essentially the same way as other transactions. Oil was recorded as leaving the joint venture inventory on the ITT date in the normal way. If several transfers were agreed within the space of a few days, it may well have been the case that Mr. Heuzé and Mr. Kilakos agreed that a single parcel should be blended to cover all of them, but even if they did not, MTI could properly regard itself as having Glencore's authority to produce the required product by blending, just as it did in the case of cargo sales. Blending generally took place a few days after the last of the relevant in-tank transfers had been agreed and Mrs. Gene informed Mr. Hawkins of the components used a few days later after receiving that information from the master of the Metrotank. The addition to stock was entered on that date as it would have been if it had been discharged into storage in the ordinary way. All this happened regardless of the absence of a letter of credit or stock transfer certificate. That may have other implications, but again, there is nothing in the way in which the parties dealt with ITT blends which supports the conclusion that MTI were given a general discretion to blend oil held for the account of the joint venture.
116. Blending was also carried out from time to time to correct overdrawing on the joint venture stocks giving rise to negative inventory. Thus on 3^rd March 1996 and again in July 1996 Mrs. Gene informed Mr. Hawkins of blends made to correct the inventory. The telex sent by Mrs. Gene to Mr. Hawkins on 19^th July referred to "Quantities/grades used for blending 380cst on 19/7/96". That did not elicit any response from Mr. Hawkins who made the appropriate entry in his spreadsheet without demur even though it must have been apparent that blending of that parcel had already taken place. I do not find that surprising given the way in which transactions of this kind were recorded. Mr. Heuzé said that he and Mr. Kilakos would discuss the shortages in the inventory and the way in which they could be made good by blending other grades which were available for the purpose. This evidence makes it very difficult to accept that Glencore was unaware that in these cases MTI was blending oil prior to the opening of a letter of credit, but it does nothing to support the proposition that MTI had a wider discretion to blend joint venture oil at will.
117. Quite apart from that, the records kept by MTI itself and the evidence of the way in which it conducted its operations in Fujairah do not bear out the suggestion that Mr. Kilakos understood MTI to be entitled to blend joint venture oil at will or that it did so. The documents as well as the evidence of the witnesses all point clearly to the conclusion that the joint venture was conducted throughout on the understanding that oil held by MTI in storage would be kept segregated by grade. Not only was that commercially necessary in order to keep reliable records of what was held in store, it was a fundamental part of the original arrangement and there is no suggestion that the parties ever agreed to depart from it. That is not to say, however, that from time to time MTI did not discharge parcels of different grades into the same tank. The evidence of Capt. Kapsomenakis shows that from time to time different grades of oil were discharged into the same tank by way of pre-blending, but only when operations to produce a finished blend were imminent. His evidence viewed as a whole does not in my view support any other conclusion. I can well understand that if blending operations are imminent it might be sensible to "store" two components in the same tank pending the full blending operation, but in my view that is properly to be regarded as a preliminary stage in the blending process rather than as a form of storage. It should be borne in mind that although for convenience of administration the parties continued to treat each cargo brought into Fujairah as retaining its original identity, that identity in fact became merged in the bulk immediately on its delivery to MTI. It does not follow, therefore, that the use of a particular cargo immediately on its arrival for the purposes of pre-blending necessarily involved a wrongful use by MTI of Glencore's property. Provided MTI continued to hold sufficient oil of the relevant grade in store to satisfy Glencore's entitlement, it was entitled to make use of oil newly delivered into storage as oil which it held for its own account.
118. In summary, I do not think that the way in which blending was carried out during the period of JV1 supports the conclusion that Glencore allowed MTI a general discretion to blend oil held on behalf of the joint venture, or indeed that Mr. Kilakos understood that to be the case.

(v) False discharge telexes

119. It was the practice of MTI to send Glencore telex confirmation of the discharge of cargoes brought into Fujairah under the joint venture agreements. It has subsequently become clear that on a number of occasions telexes were sent by Capt. Margaritis from MTI's Athens office to Glencore confirming the discharge of cargoes which had not in fact been discharged in whole or in part at Fujairah but had remained on board the vessel and had been carried on to another destination. In short, the messages were false.
120. Mr. Schaff submitted that these messages had been sent in order to deceive Glencore into thinking that the cargo had been discharged and was being held in store when that was not the case and that the need to do so arose from the fact that MTI was well aware that it was not entitled to dispose of oil brought into Fujairah from the moment of its arrival.
121. Discharge telexes assumed a particular importance from about April 1995 when the parties agreed to dispense with the use of inspectors to record the discharge of oil at Fujairah. They decided instead to accept the bill of lading figures as correct and for administrative purposes it was agreed that MTI would send Glencore a discharge telex confirming the receipt of the cargo. Capt. Margaritis who was responsible for sending these telexes was not called to give evidence. He provided a written statement to the solicitors representing Stanley Shipping and the owners of the Cherry (though not, it appears to MTI), but was apparently unwilling to give evidence in person. I have, of course read and considered what he has to say, but I have to bear in mind that his evidence has not been tested by cross-examination.
122. Capt. Margaritis worked in the Athens office under the general direction of Mrs. Gene. He said that he had been told by Mrs. Gene that an agreement had been made with Glencore that each vessel would be recorded as having completed discharge within 36 hours so that the voyage could be treated as having been completed and the freight paid. He therefore sent a fictitious telex purporting to record the details of the vessel's arrival and discharging operations for what in his view were essentially administrative reasons. He said he believed that Mr. Hawkins, to whom these messages were sent, was aware that they did not record the true position.
123. Mrs. Gene and Mr. Kilakos did give evidence about this matter and both of them gave an explanation which coincided at some points with that given by Capt. Margaritis. Mrs. Gene said that it had been agreed that vessels bringing oil to Fujairah would be sub-chartered by MTI to Glencore and would not incur demurrage after the standard time of 36 hours laytime for discharging which would otherwise have been charged as a cost to the joint venture. She said that a discharging telex confirming the completion of the voyage had to be sent to enable MTI to collect the freight from Glencore's accounting department and that the first time a vessel failed to complete discharging within the prescribed time she instructed Capt. Margaritis "to complete the voyage" and left him to take the appropriate steps. She said that she had not instructed him what to send and that on subsequent occasions he had acted without further reference to her. She assumed that he had simply taken a previous genuine message and modified it to meet the needs of the case. Mr. Kilakos also said that a telex had to be sent to terminate the voyage, and he seemed to suggest that it was necessary that it recorded that the vessel had finished discharging 42 hours after giving notice of readiness.
124. I found the evidence of Mrs. Gene and Mr. Kilakos on this point most unsatisfactory. There was no reason on the face of it why a message sent to record the completion of the voyage should not have been couched in simple terms – as indeed it was in one case. There was no need to compose an elaborate statement of facts containing false information which was liable to, and I am satisfied did on at least some occasions, mislead Glencore's operations department. I also find it difficult to accept that neither Mrs. Gene nor Mr. Kilakos was aware of the terms in which these discharge telexes were being sent. Capt. Margaritis may have been originally responsible for the form of them, but I am unable to accept that Mrs. Gene did not realise at some point what was going on, and if she knew, I feel confident that she would have drawn it to the attention of Mr. Kilakos.
125. One of the curious features of the case, however, is the fact that some false discharge telexes were sent in circumstances where they cannot have been intended to deceive Glencore, if only because to MTI's knowledge Glencore was provided with the correct information. I am grateful to Mr. Goldstone for drawing seven such examples to my attention. In one case involving the vessel Epic the whole of the cargo was discharged at Fujairah but not in the manner described in the discharge telex. In another case involving the Shoko Glencore bought part of the cargo from MTI and retained it on board for blending to provide a cargo for sale to a third party. The surveyors SGS who had been appointed by Glencore reported on the operations involving the Shoko giving information which was clearly at variance with the discharging telex sent by Capt. Margaritis. In these and other cases it can be seen that information contained in the discharge telexes was contradicted by other information of a reliable nature which MTI knew was available to Glencore.
126. When this point was put to Mr. Hawkins he said that he received a very large number of messages every day. Many he scarcely read at all and he certainly did not spend time comparing them to see if some of them contained false information. I entirely accept his evidence on that point and would have been surprised if he had said otherwise. The opportunity undoubtedly existed, therefore, for these telexes to deceive Glencore. Nonetheless, if the question is whether they were sent with intent to deceive, the matter has to be examined from MTI's point of view. If false messages were sent in circumstances where there was, to the knowledge of MTI, no possibility of any relevant deception, that does suggest that there was some other motive behind them. Although MTI's conduct in sending false telexes of this kind was deplorable and dangerous, I am not persuaded that it points very clearly to the conclusion that MTI was aware that it was not entitled to dispose of oil immediately on its arrival at Fujairah.

The Insurers' case – 'The Third Way'

127. I come now to one of the more difficult issues in the case, namely, whether, as Mr. Cooke submitted, it is possible to infer from the way in which the parties conducted the business of the joint venture prior to June 1996 that they had agreed that MTI should have an unfettered right to dispose of oil sold to it by in-tank transfer as from the ITT date, that is, the date on which the stock transfer was deemed to have taken place for payment purposes.

(i) The ITT contract

128. It is convenient to begin the discussion of this issue by considering the form of ITT contract which the parties routinely used to cover the sale of individual parcels of oil by Glencore to MTI. As I have already said, the form of this contract had become broadly settled by about June 1994. It confirmed a contract of sale on the following terms: (i) delivery to be made by stock transfer, that is, by a change in the nature of MTI's possession; (ii) stock transfer to take place on the seller's receipt of a letter of credit covering the payment of the price; (iii) payment to be made 45 days after the date of the stock transfer against an invoice and stock transfer certificate confirming the passing of property; (iv) for payment purposes only (i.e. for the purposes of calculating the 45 day credit period) stock transfer to be deemed to have taken place on a specified date; and (v) risk to pass on stock transfer, but title to pass only on delivery of the stock transfer certificate.
129. Despite the fact that the contract provided for delivery and title to pass at different times, a contract in this form is commercially satisfactory because delivery does not take place until a letter of credit has been opened and that gives the seller sufficient assurance that he will be paid. However, the very fact that the contract is structured in this way tends to emphasise the importance attaching to delivery and in the event it is the effect which the parties ascribed to the ITT date which forms the basis for the insurers' argument.

(ii) Glencore's treatment of the ITT contracts

130. Although there was no separate joint venture vehicle, both parties kept records on the basis that the joint venture had an independent existence. Each of them therefore recorded oil as being held by MTI to the order of the joint venture and as being transferred from the joint venture to the purchaser (whether MTI, Glencore or a third party) when stock was withdrawn. Mr. Cooke relied on an impressive array of documents generated by Glencore during the course of the joint venture as showing that all those who were involved in administering the joint venture arrangements on its behalf treated ITT contracts with MTI as having the effect of releasing the oil in question to MTI on the ITT date. In other words, Glencore, he submitted, treated oil sold to MTI by in-tank transfer as being fully at the disposal of MTI as from that time.
131. The primary record kept by Glencore was Mr. Hawkins' spreadsheet in which he entered each cargo or parcel of oil entering or leaving the joint venture inventory, grade by grade. When a cargo of oil was brought into Fujairah for the joint venture it was entered in the spreadsheet as having been "sold" by Glencore to the joint venture at the price at which it had been bought by Glencore. This was used to establish the base price for the calculation of profit and loss. When a cargo was supplied under the joint venture to a third party it was entered in the spreadsheet as "sold" by the joint venture to that third party. In fact, of course, the oil was sold under a contract entered into by Glencore or MTI, and if it was sold to the third party by MTI, a corresponding contract of sale was made between Glencore and MTI. Much of the oil brought into Fujairah under the joint venture during 1995 and the first half of 1996 was sold to MTI by in-tank transfers. These were entered in the spreadsheet as tank transfers from the joint venture to MTI taking effect on the ITT date. Apart from the spreadsheet, Glencore kept no other comprehensive record of oil held by MTI for its account. It did not, for example, keep any corresponding record of the amount of oil sold but awaiting letter of credit or of parcels of oil awaiting the issue of a stock transfer certificate. The spreadsheet was used as the primary source of information relating to the movement of oil held at Fujairah for the benefit of the joint venture.
132. At quarterly intervals during this period the parties reconciled their positions and calculated the amount due one way or the other under the joint venture agreement. For this purpose Glencore drew up a detailed statement of the transactions taking place during the previous quarter showing the costs incurred in providing cargo for the account of the joint venture and the amounts due in respect of sales by the joint venture. In-tank transfers were treated as occurring within the relevant quarter for the purposes of profit and loss calculations if, but only if, the ITT date fell within that quarter. No account was taken of the date of the ITT contract itself, the date on which the letter of credit was opened or the date on which the stock transfer certificate was issued. This almost certainly reflects the fact that from an early stage the spreadsheet was treated by the operations department within Glencore, and indeed others, as the definitive source of information concerning transactions carried out under the joint venture arrangements.
133. Other documents generated by Glencore present a similar picture. These include its internal accounts and certain contemporaneous documents in which oil is described as having been "transferred" to MTI on the ITT date. Glencore's accounting manual called for goods in which it retained risk and the rewards of ownership to be included in its inventory of stocks for valuation purposes and for sums due under contracts of sale to be entered as receivables in those cases where it no longer retained risk or the rewards of ownership. It was common ground, however, that the accounts treated oil sold under ITT contracts as having left the inventory on the ITT date and the sums due from MTI in respect of that oil as receivables as from that date. This was the result of employees in the accounts department treating the information derived from Mr. Hawkins' spreadsheet as accurately reflecting movements in the joint venture stock. Mr. Cooke submitted in the light of this evidence that there must have been a considered decision by Glencore's accounts department to treat joint venture stocks in that way and that it must have been taken following discussions with Mr. Heuzé and Mr. Hawkins about the nature of the agreement with MTI. However, in the light of the evidence of Glencore's witnesses, particularly that of Miss Lorimer who explained how the information had originally been collated and Mr. Tawana who was responsible for collecting and transmitting the relevant information from the operations department from August 1997 onwards, I am not persuaded that that is really what took place. I am satisfied that from an early stage the accounts department simply assumed that the spreadsheet represented an accurate record of the movement of stocks into and out of the joint venture inventory and that no one stopped to consider in any detail the exact status of oil sold under an ITT contract pending the opening of a letter of credit or the issue of a stock transfer certificate.
134. Mr. Cooke also relied on a file note apparently produced by Glencore's insurance brokers, Lloyd Thompson, recording a meeting with Mr. Warren and Mr. Gibson of Glencore's insurance department in May 1997 relating to the renewal of its cover on oil held in the floating storage. The note records Mr. Warren as having said that 90-95% of the oil then going into store was blended and off-loaded within 30 days. In this context "off-loaded" must mean physical movement out of storage. The insurers have put the authenticity of this document in issue and that is a matter which can only be determined in the light of such evidence as may be called in a later phase of this litigation when the court deals with the insurance issues. For present purposes I can only proceed on the assumption that it may not be in all respects genuine. It was common ground, however, that by 1997 oil sold to MTI would not normally be the subject of a stock transfer certificate within 30 days of arrival at Fujairah. Indeed, during the latter part of 1996 and the first half of 1997 a stock transfer certificate was not generally issued in less than about 50 days. On the other hand, during the same period the time between arrival and ITT date was usually less than 30 days, sometimes much less.
135. Mr. Cooke submitted that whether or not this note is authentic in the sense that it records discussions between Glencore and Lloyd Thompson on 27^th May 1997, there is no reason to doubt that it reflects the understanding of Glencore's insurance department of the manner in which operations were being carried out at Fujairah, an understanding that can only have been obtained from Mr. Hawkins. Mr. Hawkins accepted that he was approached by the insurance department from time to time for information, but he denied telling Mr. Warren that most cargoes were blended and off-loaded within 30 days. His recollection was of being asked what proportion of cargoes coming into Fujairah under the joint venture were insured by Glencore for the voyage, as reflected in a manuscript note made by Mr. Warren at around the same time.
136. Although the thrust of the enquiry in either case relates to the 30 days' free insurance after discharge, there is a significant difference between the two enquiries. Whatever the origin of the brokers' note, the reference to blending and off-loading must have had its origin within Glencore, and must therefore have been derived from something said by Mr.Hawkins. It is possible that Mr. Warren misunderstood what was said to him, but I think it more likely that Mr. Hawkins did give him the impression that cargoes were being turned round within the 30 day period of free insurance, which of course was consistent with his spreadsheet.

(iii) MTI's treatment of the ITT contracts

137. MTI kept a number of records relating to oil held in the floating storage facility. They fall into three groups. The first consists of two corresponding sets of manuscript ledgers, one of which was kept in the Athens office and one on board the Metrotank, in which MTI recorded movements of stock which it held as available for its own use. They record the source, date and amounts of additions to stock and the dates and recipients of oil leaving stock. The second group consists of records of oil held for the account of the joint venture. A manuscript ledger of a kind broadly similar to those recording MTI's own stock movements was kept in the Athens office in respect of joint venture stock. A corresponding computerised record was kept on board the Metrotank. In the case of in-tank transfers all these ledgers recorded oil as becoming available to MTI on the ITT date. In addition to these there was the computerised spreadsheet originally introduced by Mr. Griffin and subsequently maintained by Mrs. Gene. This also treated oil sold by the joint venture to MTI as leaving the joint venture inventory on the ITT date. The third group consisted of a series of "daily reports" produced for Mr. Kilakos by the Athens office showing, among other things, MTI's commitments and the oil available to meet them.
138. According to Mr. Kilakos daily reports were first produced in the late 1980's or early 1990's, although the first examples produced in evidence date from March 1996. They were produced by the office in Athens on the basis of the manuscript ledgers and took the form of printed tables setting out, among other things, quantities of oil in inventory by grade, quantities sold and quantities shortly coming into Fujairah for the account of the joint venture. Manuscript amendments made during the day to record changes in the position were reflected in the next day's printed sheet. The first table in the daily report showed the quantities of oil of different grades available on the storage vessels – in other words the oil currently available to MTI for fulfilling its commitments. One important aspect of these sheets for present purposes is that they show oil in the joint venture inventory becoming available to MTI on the ITT date rather than the date on which it was discharged into storage or the date on which a letter of credit was opened or a stock transfer certificate issued. From time to time the quantity of oil on storage vessels is shown in the daily reports as a negative figure. According to Mr. Kilakos that indicated that oil had been drawn by MTI from the joint venture stock before any ITT contract had been issued to cover it.
139. Another section in the daily report recorded what was described as "cargo in storage". This took the form of a table in which there were entered details of cargoes delivered into store at Fujairah in respect of which no contract of sale had yet been issued. When a parcel of oil was sold to MTI it was deducted from the oil shown in this section and the relevant quantity was added to another table headed "Remarks included in cargo available – no L/C". As this heading suggests, the oil shown in this table had been sold by Glencore to MTI by in-tank transfer, but no letter of credit had yet been opened in respect of the price. Against each parcel there was recorded the price and the relevant ITT date. One can see from these sheets, therefore, that MTI was treating cargo as available to it for disposal from the ITT date, but not before.
140. Two other kinds of document routinely produced by MTI in connection with the operation of the joint venture should also be mentioned. First, the storage invoices. For reasons I have already mentioned, from about the autumn of 1994 MTI was content to charge a flat rate fee of $1 per ton in respect of all cargoes brought into Fujairah except for a few products for which a higher rate had been agreed. That applied mainly to various types of cutterstock which were normally used in relatively small quantities over longer periods of time. For the purposes of calculating storage charges MTI treated parcels of cutterstock sold by in-tank transfer as having left storage on the ITT date.
141. Secondly, there are the telexes sent by MTI to Glencore from time to time advising it of components used to manufacture blended products, the so-called "blending telexes" to which I have already referred. It was not unusual at that time for several in-tank transfers of the same grade of fuel to be made on the same or successive days and for a single blending telex to be sent some days later covering all the parcels. Mr. Hawkins recorded these transactions in his spreadsheet as transfers of the blended product from the joint venture to MTI taking place on the ITT date, the movements brought about by the blends themselves being entered on the date on which he received the advice. These were recorded as reductions in the stock of the components and a corresponding increase in the stock of the blended product.
142. Other documents generated by MTI tell a similar story. Included in the documents before the court are letters passing between the office of BTCL which handled MTI's account and Credit Lyonnais seeking authority to open letters of credit covering the price due from MTI under ITT contracts for the purchase of fuel oil. One can see from these that the bank had been told by MTI that delivery of the oil had taken place many days earlier on a date which was in fact the relevant ITT date. Similarly, in its own internal accounts MTI treated the ITT date as the date on which oil passed from the joint venture inventory to MTI's inventory.

(iv) Glencore's knowledge of MTI's storage capacity

143. Mr. Cooke submitted that Glencore was always well aware that the capacity of the storage vessels at Fujairah was not sufficient to contain the quantity of oil awaiting letter of credit and stock transfer certificate. Accordingly, it must have known about and consented to MTI's disposing of oil prior to the issue of a letter of credit. For this reason the extent of Glencore's knowledge of the capacity of the storage facility at any given time is of relevance both to the issues which arise in Phase 2 and to issues between Glencore and the insurers which will fall for determination in a later Phase of this litigation.
144. Apart from the Metrotank the identity of the vessels making up the storage facility varied from time to time. Moreover, the nature of MTI's operations was such that vessels which were primarily used for the carriage of oil to and from Fujairah could be, and on occasions were, used for the purposes of storage to meet short term operational requirements. It is sometimes difficult, therefore, to identify with any precision which vessels can properly be said to have been part of the storage facility at any given time. That is not a matter on which it is necessary or appropriate to make specific findings at this stage. Some vessels, on the other hand, can be seen to have been an established part of the storage facility at different times. These include, for example, the Mount Athos, which was used for this purpose at different times during the joint venture, and the Violet, Nafkratis, Athenian Splendour and Sea Giant, all of which were mentioned in Mr. Heuzé's notebook in January 1998. From the latter part of 1994 Mr. Heuzé was in frequent conversation with Mr. Kilakos and I have little doubt that as a result of their discussions he was broadly aware of the identity of the main storage vessels being used by MTI at any given time. A relationship of trust began to develop between them from an early stage, however, and deepened as time went on. Partly for that reason I do not think that Mr. Heuzé concerned himself with the details of the floating storage or that he interested himself in the number or identity of smaller vessels which may have been used for storage purposes from time to time. It was inherent in the nature of the facility that MTI could expand or reduce its capacity to meet demand and Mr. Heuzé was undoubtedly aware of that. To that extent he must have known that additional vessels might be used for storage purposes from time to time. It is possible that on various occasions Mr. Heuzé became aware that a particular vessel was about to be used, or was being used, for storage purposes, but that is not a matter that was explored with him and I am not in a position to make any findings about it.
145. Mr. Hawkins was also aware of the nature of the floating storage facility and of the fact that MTI could, and would, increase or reduce its capacity to meet demand. Both he and Mr. Heuzé were aware that the Mount Athos was leaving the facility in May 1994 and were also aware of the vessel's return in January 1995. Mr. Hawkins had frequent occasion to speak to MTI's loading master, Capt. Kokolakis, about arrangements for the loading and discharging of cargoes and he also received routine telex confirmation of cargoes delivered out of storage which identified the vessels from which they had been loaded. It seems clear, therefore, that if he had chosen to do so he could have identified many of the vessels being used for storage purposes at any given time. However, a more sustained analysis of that information would have been necessary to enable him to work out which of them was being used on a long term basis and I am satisfied that he did not in fact carry that out.
146. With the benefit of hindsight one can see that the capacity of the floating storage facility, particularly during the latter months of 1997, fell far short of what would have been necessary to contain all the oil in respect of which no stock transfer certificate had been issued. It does not follow, however, that Glencore must necessarily have been aware of that fact. The fact that MTI could easily obtain additional capacity by chartering some of the many vessels waiting for employment in and around Fujairah means that there was always an element of uncertainty about the capacity of the facility. Mr. Heuzé said that he did not take the trouble to work out the storage capacity available to MTI and was not aware that it fell substantially short of what would have been necessary to contain all the oil which he believed should be held to Glencore's order. I accept what he said about that. Although his lack of interest in these matters may at first sight seem surprising, it is necessary to bear in mind, first, that he was a trader, not an operations man, and secondly, that by the beginning of 1997 a considerable degree of trust had developed between him and Mr. Kilakos. The fact that Glencore as an organisation did not take steps to monitor its stocks at Fujairah otherwise than through the medium of Mr. Hawkins' spreadsheet is beside the point. That sort of exercise might have been carried out by the operations or accounts department, but is not one which Mr. Heuzé as a trader would normally be expected to do. Nor is it one which anyone asked Mr. Hawkins to do. In the circumstances I am satisfied that neither Mr. Heuzé nor Mr. Hawkins was conscious of the extent to which the capacity of the storage facility fell short of what they would have expected. The insurers' argument does not therefore derive any support from this quarter.

(v) The parties' individual views

147. The body of contemporaneous documents to which I have referred leaves me in no doubt that each party, for whatever reasons, conducted the business of the joint venture on the basis that a genuine transfer to MTI of some kind occurred on the ITT date. At the time the parties agreed an in-tank transfer the oil was already in the physical possession of MTI and there is nothing in the evidence to suggest that either party understood or intended that at the time of stock transfer some change of merely theoretical nature would occur in the quality of MTI's possession. One can see from the very fact that in the standard form of ITT contract itself stock transfer was linked to the opening of the letter of credit that it was understood to involve something much more significant. It is difficult to see what that could be other than a right to dispose of the goods, and indeed, Mr. Heuzé made it clear on a number of occasions that he was quite happy for MTI to make use of the oil once a letter of credit had been put in place. This helps to explain why Mr. Palacios was puzzled when shortly after the collapse of MTI in February 1998 he tried to understand how the ITT contract worked. He found it difficult to understand how there could be a stock transfer at one point and a stock transfer certificate taking effect at a later date. Although someone within Glencore, probably, I think, Mr. Heuzé, wanted to ensure that Glencore formally retained title until payment had been received, I have little doubt that it was generally understood that in principle once the stock transfer had occurred MTI was entitled to dispose of the oil. Mr. Heuzé came close to admitting that, if he did not actually do so, and the language used in Glencore's documents reflects that understanding.
148. What seems to have happened is that from a very early stage Mr. Hawkins treated the ITT date, rather than the date on which the letter of credit was opened, as the effective date of the stock transfer and therefore of delivery. (This was the equivalent of the date on which instructions were given to the master of the Mount Athos to release cargo.) Consistently with that he equated the ITT date in the case of in-tank transfers with the bill of lading date (i.e. the date of physical delivery) in the case of deliveries of cargoes shipped out of Fujairah and the date of receipt of the blending advice in the case of physical blends. That was the basis on which the spreadsheet was kept and everyone else simply relied on the information in the spreadsheet as correctly reflecting the position. As far as MTI is concerned, it acted on the basis that the ITT date was the date of delivery and that delivery gave it a free hand to deal with the oil.

(vi) Communications between the parties

149. For these reasons I am satisfied, therefore, that during the period of JV1, certainly from the early part of 1995 if not before, both parties treated the ITT date in their records as the point at which MTI became entitled to dispose of the oil covered by the contract. In Glencore's case that was not the result of a conscious decision, but nonetheless that is the picture which a neutral observer would obtain from an examination of its records. However, that alone is not sufficient to establish the existence of an agreement between them to that effect, or even to establish that Glencore gave its approval to MTI's disposing of the oil at the ITT date. For that it is necessary to show that there was sufficient communication between the parties of their intention to treat oil sold by in-tank transfer as having been delivered and released to MTI on the ITT date rather than at the time of opening a letter of credit.
150. Increasingly during 1994 Mr. Heuzé communicated with Mr. Kilakos directly rather than through Mr. Griffin. From the latter part of 1994 onwards they were in regular communication over the sale of bunkers for supplying the local market and as the scope of the joint venture business increased during 1995 so also did the frequency of their conversations. Among other things they discussed the purchase of oil to be brought into Fujairah for the joint venture and the marketing of cargoes. When Mr. Heuzé was away from the office he would carry a copy of Mr. Hawkins' spreadsheet on his portable computer. There was also frequent communication between Mr. Hawkins and Mrs. Gene who between them handled the bulk of the administration of the joint venture business and also between Mr. Hawkins and Capt. Kokolakis, relating to arrangements for the physical movement of oil. It is fair to say, however, that none of the witnesses gave evidence of any conversation in which the operation of the ITT contracts in general, or the significance of the ITT date in particular, was expressly discussed. In order to determine whether the parties reached any agreement about that it is therefore necessary to look at the documents which passed between them and the context in which that occurred.
151. Sales by in-tank transfer were concluded orally between Mr. Heuzé and Mr. Kilakos and recorded by Mr. Heuzé in a deal ticket. Among other essential terms they agreed was the ITT date. The deal ticket was purely an internal document and was used by Mr. Hawkins as the basis for drafting the ITT contract confirming the sale which he sent to Mrs. Gene by telex a few days later. Although the ITT contract in its final form is not a model of clarity, I do not accept Mr. Cooke's submission that it is internally inconsistent, nor, for the reasons I have given, do I think it inherently uncommercial. In my view it is reasonably clear as a matter of construction that it provides for delivery by stock transfer to take place on receipt of a letter of credit.
152. The principal documents which passed between the parties relating to the administration of the joint venture were the spreadsheets kept by Mr. Hawkins and Mrs. Gene. By the early part of 1995 it had become the established practice of Mr. Hawkins, Mr. Griffin and Mrs. Gene to send each other from time to time copies of the sheets covering recent transactions to enable them to reconcile any differences. In addition general reconciliations were carried out at the end of 1995 and 1996 to agree the quantities to be carried forward into the spreadsheet for the following year. There was a difference between the parties as to the frequency with which spreadsheets were exchanged: Mr. Hawkins said that it did not happen very often; Mrs. Gene, on the other hand, said exchanges took place every week and sometimes more often. This difference is not of great significance, however, since it is clear from the contemporary documents that spreadsheets were exchanged regularly at intervals of no more than three to four weeks. Any discrepancies were discussed and resolved by telephone or in correspondence. Other documents which passed between the parties, such as the monthly invoices for storage charges and the quarterly profit and loss reconciliation statements, were based on information derived from the spreadsheets insofar as the date of delivery of the oil was relevant.
153. An exchange of information of a different kind occurred around the end of 1994. On 23^rd December 1994 Glencore wrote to MTI asking it to provide confirmation to its auditors of the amount of oil held by MTI in stock for Glencore at 31^st December 1994. On 16^th March 1995 Mrs. Gene responded on behalf of MTI setting out the quantities shown in her spreadsheet as representing the joint venture stock at the year end. This answer was accepted without challenge, perhaps not surprisingly since it reflected the figures to be found in Mr. Hawkins' spreadsheet which had in turn been used by Glencore's accounts department to generate its own records.
154. As Mr. Cooke pointed out, all those involved in this trial recognised that to a greater or lesser degree the terms of the parties' relationship could only be derived from their conduct. Glencore relied on the adoption by conduct of the central terms of the agreements made in December 1993 and May and July 1994 in relation to the original cargoes of straight run fuel oil. MTI relied on the creation by a combination of oral communications and conduct of a joint venture agreement which was to apply to a new range of cargoes not covered by the earlier agreements. Mr. Cooke submitted, however, that in these circumstances, and particularly in the absence of any written over-arching contract, it was necessary to examine the parties' conduct as a whole to determine the terms on which they did business. He therefore challenged Mr. Schaff's submission that the present enquiry should be confined to the question whether the parties had agreed at some identifiable point to depart from an established understanding that Glencore would only sell oil to MTI on secured terms. He submitted that it was necessary to look at the evidence as a whole in order to determine at what stage in a routine sale by in-tank transfer the parties had agreed that MTI should become entitled to make use of the oil.
155. The creation of contractual relations depends on an intention on the part of each party, manifested to the other, to assume contractual obligations to the other: see Harvela Ltd v Royal Trust Co. [1986] AC 207, 226 per Lord Diplock. It has long been recognised that conduct may play an important part in determining whether an intention of that kind has been manifested by one or other party, but the question most commonly arises in cases where the terms of the proposed contract have been communicated by one party to the other, either orally or in writing. The issue then is whether, by conducting himself in a particular manner, the other party has demonstrated an intention to be bound on those terms. The present case, however, has two rather unusual features. First, each sale of oil by Glencore to MTI was the subject of a separate oral agreement between Mr. Heuzé and Mr. Kilakos which subsequently gave rise to a telex confirmation which contained terms which are complete in themselves and deal specifically with the date on which delivery of the goods is to take place. Secondly, any agreement that the goods were to be delivered and released to MTI on the ITT date rather than against the opening of a letter of credit is inconsistent with the terms of the telex confirmation and moreover must have been derived from the parties' conduct without being expressly defined in any way.
156. Both Glencore and MTI accepted that the individual ITT contracts were made within the framework of an over-arching joint venture agreement and Mr. Cooke therefore submitted that the court's task in this case was simply to ascertain the terms of that over-arching agreement. That is all very well, but it is impossible to ignore the fact that each ITT contract represented a self-contained contract of sale and that part of the parties' conduct included sending and receiving a formal contract confirmation on each occasion without any apparent objection. If the parties did agree that the oil was to be released to MTI on the ITT date rather than on the opening of a letter of credit, they must also have agreed by implication that any provision to the contrary in the ITT contract was to have no effect. An alternative, and perhaps better, view might be that the ITT contract did not accurately reflect the agreement between the parties.
157. In this context Mr. Schaff drew my attention to the decision of the Court of Appeal in Haryanto v E.D. & F. Man (Sugar) Ltd [1986] 2 Lloyd's Rep. 44. The issue in that case was whether two contracts for the sale of sugar, each of which had been signed by the plaintiff and was to all appearances regular and enforceable, were in fact binding upon him. The plaintiff maintained that the contracts were not intended to be binding contracts at all but were merely intended to record the terms on which the defendants would be prepared to contract with a third party. Both Staughton J. at first instance and Lloyd L.J. in the Court of Appeal referred to a passage in the judgment of Diplock L.J. in Garnac Grain Co. Inc v H.M.F. Faure & Fairclough Ltd [1966] 1 Q.B. 650 at page 683 where he said this:

" . . . . . . . . unless some question of waiver or estoppel arises the contemplation or intention (unless incorporated in the contract) of the parties or either of them as to the way in which it will be performed or left unperformed does not affect their legal rights or obligations under it. To affect these it is necessary to go further and to show that the parties really made some other and different contract between them and agreed that the ostensible contract should not give rise to legally enforceable rights or liabilities."

158. The fact that each of the ITT contracts was made within the framework of the over-arching agreement known as JV1 does not seem to me to detract from the fact that each of them purported to be an independent contract of sale. In reality the over-arching agreement did little more than regulate the terms on which profits and losses arising from sales of joint venture cargoes were to be shared. As in the case of sales by Glencore to third parties on behalf of the joint venture, the individual ITT contracts themselves purported to regulate the terms on which the goods to which they related were sold. I think Mr. Schaff was right, therefore, in saying that there is a burden on the insurers to show that there was agreement between Glencore and MTI that the terms set out in the standard ITT contract should not apply.
159. Mr. Schaff objected that when Mr. Heuzé agreed to sell oil to MTI by in-tank transfer he had no intention of agreeing that MTI should be allowed to make use of that oil before it had opened a letter of credit. That may be so, but the manifestation of an intention to assume contractual obligations to which Lord Diplock referred in Harvela Ltd v Royal Trust Co. depends not on subjective intention but primarily upon the objective assessment of communications made by words or conduct: see Chitty on Contracts, 28^th ed. §2-148. This may pose particular difficulties in a case where the very existence of a contract is in issue: see Mitsui & Co. Ltd v Novorossiysk Shipping Co. (The Gudermes) [1993] 1 Lloyd's Rep. 311. However, I do not think that the present case falls into quite the same category since it is not disputed that a contract did come into existence each time Glencore agreed to sell a parcel of oil to MTI.
160. I have already found that during 1994 and the early months of 1995 MTI understood that Glencore was only willing to sell oil to MTI on letter of credit terms and that there was no subsequent agreement to trade on open account terms. Indeed, that remained the case and so throughout the course of the joint venture MTI continued to open letters of credit in respect of the price payable under these contracts. Yet it is also clear that right from the very beginning Mr. Hawkins kept his record of the joint venture inventory in a manner consistent with the delivery of oil to MTI on the ITT date. He explained that by saying that the spreadsheet was intended to show how much oil in the joint venture inventory was uncommitted for sale, but in fact it is clear that it was treated as a physical inventory in many respects. For example, when oil was sold under a contract which called for shipment ex Fujairah, it was shown as leaving the inventory on the bill of lading date (i.e. the date of physical delivery) rather than the date of the contract.
161. During 1994 there were at least two reconciliation meetings, one of which was attended by Mr. Heuzé, at which the calculation of the joint venture profit and loss was discussed by reference to statements drawn up on the same basis. Viewed objectively, the regular exchange of information in the form of inventories, reconciliation statements and storage invoices did in my judgment give rise to a clear representation on the part of Glencore that it, like MTI, regarded the ITT date as the date upon which oil sold by in-tank transfer was delivered, or in Mr. Hawkins' terms "released", to MTI. Indeed, if at the time Mr. Hawkins had been asked when oil sold under an ITT contract was transferred from the joint venture to MTI, I am confident that his answer would have been that it was transferred on the ITT date shown in the spreadsheet.
162. The exchange of documents on which Mr. Cooke relied took place mainly, though not entirely, between Mr. Hawkins on the one hand and Mr. Griffin and Mrs. Gene on the other. They were not the persons through whom the parties contracted and Mr. Schaff submitted that there was no relevant exchange of information between Mr. Heuzé and Mr. Kilakos. As I have already mentioned, there was at least one reconciliation meeting during 1994 at which Mr. Heuzé was present for which a statement was prepared on the basis that delivery occurred on the ITT date, but I do not think that the matter rests there. Mr. Heuzé naturally interested himself in the joint venture inventory; he had to, because as the trader he needed to know how much oil was at his disposal. He must have been aware that Mr. Hawkins was treating oil sold by in-tank transfer as delivered to MTI on the ITT date. He must also have been aware that that fact would be communicated to MTI when Mr. Hawkins and Mrs. Gene reconciled their positions. Whether Mr. Heuzé gave any careful consideration to the implications of acting in that way is another matter. I am not persuaded that he did. As far as one can tell, Glencore considered that it was the retention of title which provided its effective security, as indeed is borne out by the fact that a stock transfer certificate was usually issued soon after a letter of credit had been opened. The potential inconsistency between delivering oil to MTI and retaining an effective form of security does not seem to have been considered by anyone at the time.
163. Although Mrs. Gene was the point of contact for this exchange of information as far as MTI was concerned, I am satisfied that Mr. Kilakos must also have been aware of the manner in which Glencore was keeping its own inventory. He did not look at Mrs. Gene's spreadsheet frequently, but he did see it from time to time and was aware that, in common with MTI's other records, it treated oil as being transferred from the joint venture inventory to MTI on the ITT date. He was aware too that Mrs. Gene and Mr. Hawkins were reconciling their positions by exchanging inventories from time to time and is likely to have appreciated from that fact alone that Glencore was keeping its inventory in the same way. Had there been any indication that Glencore was keeping its inventory on a basis substantially different from that on which MTI was working, I am sure that he would have raised the matter with Mr. Heuzé. Although Mr. Kilakos did not concern himself frequently with storage invoices or the reconciliation procedure, he did have reason to deal with them on occasions and must have appreciated the basis on which these documents were drawn up. In these circumstances I do not think there is much doubt that he understood that Glencore, like MTI itself, was treating the ITT date as the date of delivery.
164. Glencore's representation that it regarded the ITT date as the date upon which oil sold by in-tank transfer was delivered to MTI was capable of providing an important part of the context in which Mr. Kilakos agreed to buy oil by in-tank transfer. Nonetheless, the objective approach to the ascertainment of contractual intention cannot be relied on if the person to whom the relevant representation is made knows that it is not intended to be taken at face value: see Chitty, op. cit. §§2-148 and 2-003. Mr. Schaff submitted that even if Glencore had conducted itself in such a way as would give the reasonable observer the impression that it regarded the ITT date as the date on which oil was released to MTI, MTI cannot have believed that that was really its intention given the fact that Glencore's insistence on a secured transaction had already been clearly established.
165. The positions adopted by Glencore and MTI in these proceedings were at opposite extremes, Glencore contending that MTI was not entitled to dispose of oil before a letter of credit had been opened and MTI contending that it was entitled to dispose of oil from the moment of its arrival at Fujairah. Only Mr. Cooke on behalf of the insurers, therefore, had an interest in cross-examining Mr. Kilakos on the basis that the ITT date was in fact the critical date, as reflected in the documents. When questioned about the relevance of the ITT date Mr. Kilakos went a long way towards accepting, contrary to MTI's case, that he had at the time understood it to be the date on which oil became available to MTI. Mr. Schaff quite rightly cautioned me against accepting Mr. Kilakos' evidence on this question too readily and I accept that there was some force in his criticisms of it. Nonetheless, it does clearly reflect what one finds in the contemporaneous documents and for that reason I am satisfied that it is reliable. I therefore reject the suggestion that Mr. Kilakos realised that Glencore did not intend to release oil to MTI on the ITT date.
166. In the present case Mr. Cooke submitted that it was possible as a matter of common sense to see that there was agreement between the parties concerning the date of delivery and release of oil to MTI under these ITT contracts without the need for any sophisticated analysis. He reminded me of the passage in the judgment of Hobhouse J. in General Accident Fire and Life Assurance Corporation v Tanter, The Zephyr [1984] 1 Lloyd's Rep. 58 at page 72 in which he said

"Where, as in the present case, there is a clear intent to create legal relations and the transaction or transactions are clearly of a commercial character, English law is perfectly ready to recognise the contractual relations that the parties actions so clearly intend and will not frustrate them on account of some difficulty of analysis."

As a general principle I would respectfully agree. Nonetheless, when seeking to determine whether two parties have agreed terms informally, especially when those terms are said to differ in material respects from those set out in a document which purports to record the agreement, I think one should approach the matter with some caution. In particular, the discipline of identifying offer and acceptance in the usual way is likely to provide the safest guide. Where conduct is relied on as giving rise to an agreement there is undoubtedly a danger of imputing to the actions of one or other party a greater significance than is really justified: see The Gudermes.

167. By the end of 1994 the manner in which, to the knowledge of both parties, in-tank transfers were being administered on both sides had already become an important part of the commercial context in which the joint venture arrangements subsequently developed. Through their exchange of information each had objectively represented to the other that delivery of oil sold by in-tank transfer on the usual terms would in fact take place on the ITT date, notwithstanding the terms recorded in the standard form of telex confirmation, and this provided a conventional basis for their dealings. Glencore's representation that it regarded the ITT date as the date of delivery became part of the terms on which the parties negotiated for and agreed to enter into such contracts, with the result that each subsequent contract incorporated a term to that effect. Since I am ultimately only concerned with the position in 1997 and the early part of 1998, I do not think it matters precisely when that became the established basis of their dealing, although I think it occurred at quite an early stage since the reality of the matter is that for a long time prior to June 1996 the parties had entered into contracts for sale by in-tank transfer on that understanding. In any event it is possible to say with confidence that by June 1996 when the transition to JV2 occurred it represented the established basis on which the parties did business together. I do not think that this poses any real difficulty of analysis because Glencore's representation that oil sold by in-tank transfer would be treated as delivered to MTI on the ITT date provided the unspoken basis on which all such contracts were made. An objective observer standing in the shoes of Mr. Heuzé and knowing what he knew about the commercial background to that request would have realised that. To hold that the parties nonetheless intended to contract on terms that delivery would not occur until a letter of credit had been opened would be to ignore commercial reality.

Subsequent developments: JV2, June 1996 - February 1998

168. The joint venture had proved profitable to both parties during part of 1995, but the position began to change during the first half of 1996 and by June of that year Glencore had reached the conclusion that it was not generating sufficient profit to justify the volume of oil being handled and the degree of market exposure involved. This led to discussions over a number of days between Mr. Heuzé and Mr. Kilakos in June 1996 while they were both staying in Singapore. There was some dispute between the parties about the circumstances leading up to these discussions, to which I shall return. On any view it was agreed on all sides, however, that the discussions were very informal in nature, much of them being conducted in and around the swimming pool. No note was made on either side, either of the discussions themselves or, more surprisingly, of their outcome. Nonetheless, it is common ground that they culminated in the agreement under which the parties conducted most of their business together over the following eighteen months.
169. The informality of the discussions and the absence of any confirmatory telex or other record of what was agreed has resulted in there being very little direct evidence of what passed between Mr. Heuzé and Mr. Kilakos on that occasion. Neither claimed to have a good recollection of any specific conversations, but fortunately there was a large measure of agreement about the main topics raised between them and the outcome of their discussions. Glencore's primary concern was to limit the extent of its market exposure and to improve the profitability of their co-operation, either by reducing the volume of oil handled by the joint venture to the amount needed by MTI to meet its requirements for bunkers, or by confining its own interest to a modest fixed profit on all oil purchased for the joint venture, leaving the marketing entirely to MTI. In the end it was agreed that, except in those cases where they specifically agreed to share in marketing profits and losses (which would continue to be handled under the existing JV1 arrangements), all oil bought for the joint venture in the future should be sold by Glencore to MTI at a price which reflected the full cost of acquisition and storage, including financing costs in the form of interest, plus a fixed joint venture profit of $1.50 per ton which would be shared equally between the parties. They agreed that MTI should continue to receive a storage fee of $1 per ton, but since that formed part of the total acquisition cost it did not represent a net payment to MTI. The effect of these arrangements was that Glencore received a profit of $0.75 per ton on all oil brought into Fujairah under the joint venture and MTI assumed complete responsibility for buying it from Glencore and marketing it to third parties.
170. Mr. Crane submitted that in the course of their discussions in Singapore Mr. Kilakos had made it clear to Mr. Heuzé in one way or another that if there were to be a new arrangement under which MTI purchased all the oil brought into Fujairah it would need to assume complete control over that oil from the moment of its arrival. In any event, he submitted, that was essential to the nature of the proposal and must have been understood as such by Mr. Heuzé.
171. It is certainly true that the agreement made at Singapore brought about a fundamental change in the nature of the parties' relationship. Given the nature of JV2, it now seems a little curious that they chose to structure it as a joint venture and in many ways continued to regard it as such. Mr. Heuzé agreed that that was illogical and Mr. Kilakos did not find it easy to explain why he was willing to pay Glencore $1.50 per ton in the first place only to recover $0.75 at a later date. I think the explanation probably lies in the fact that both Mr. Heuzé and Mr. Kilakos preferred to build on their existing arrangements rather than start all over again. That was entirely consistent with the way in which they had conducted themselves in the past, and having seen Mr. Heuzé and Mr. Kilakos give evidence I do not find it at all remarkable. The fact that they did choose to pursue the relationship in that way is, however, of significance in itself because it tends to support the conclusion that in other respects they intended it to remain substantially unchanged.

Was MTI entitled to dispose of the oil as soon as it reached Fujairah?

172. Mr. Kilakos maintained that he believed that MTI was entitled to dispose of the oil brought into Fujairah under JV2 from the moment it was discharged into storage. He said it was inherent in the agreement that MTI should have sole responsibility for marketing it, but in support of that contention he sought to set the discussions in Singapore between himself and Mr. Heuzé in June 1996 in the context of an impending purchase by him of MTI from Mr. Mavrakakis. He said that he had explained to Mr. Heuzé his need for additional financial support and that Glencore had agreed to provide it by allowing MTI to have complete control over all oil brought into Fujairah under JV2 from the moment of its arrival, thereby enabling it to sell it before it was obliged to open a letter of credit. Mr. Heuzé denied that there had been any agreement of this kind, and I have no hesitation in accepting his evidence on this question. The documents show that the agreement between Mr. Kilakos and Mr. Mavrakakis for the purchase of MTI did not come into existence until December 1996 and it was probably not until early 1997 that Mr. Kilakos first mentioned the matter to Mr. Heuzé. It is true that Glencore was willing to lend financial support to Mr. Kilakos and the Metro companies at that time, as appears from the arrangements which they entered into for the funding of an expansion to the refinery in Fujairah operated by Metro Oil Corporation ("MOC"), but it was only willing to do so in a formal manner and against satisfactory security. Mr. Schaff submitted that the documentary evidence relating to the financial affairs of MTI and MOC during 1997 show clearly that very substantial sums were transferred from MTI to MOC during that period which had been derived from the sale of oil forming part of the joint venture stock. He also submitted that substantial quantities of 380cst fuel oil were drawn from the joint venture stocks to enable MOC to meet its obligations under a processing agreement with Texaco. All this would go a long way towards explaining the situation which later came to light when MTI collapsed in February 1998, but it is unnecessary for me to make findings in relation to either allegation in order to determine the issues which arise in this Phase of the litigation and I do not therefore think it desirable to lengthen this judgment by dealing with them in the detail that would be required in order to do so.
173. I am unable to accept either that Mr. Kilakos asked Mr. Heuzé to give him the freedom to dispose of oil as soon as it reached Fujairah and to account for it in arrears, or that, if he had done so, Mr. Heuzé would have agreed to such a request. Whatever may have been Glencore's perception of the arrangements in place under JV1, I am quite sure that Mr. Heuzé would not have agreed to release oil to MTI without qualification from the very moment of its arrival at Fujairah since such an agreement would quite clearly have deprived Glencore of the benefit of any security. But in any event, it was common ground that there was no express discussion of the ITT contracts, the storage arrangements or any other aspect of the existing machinery of JV1. Nor was there any discussion of more practical matters such as blending or the transhipment of incoming cargoes. In these circumstances I am satisfied that there was no agreement to alter the manner in which cargoes were bought for the joint venture and sold to MTI by in-tank transfer and that both parties proceeded on the understanding that the existing system should remain essentially unchanged, as indeed it did, at least to begin with. Mrs. Gene said that she had learned of the existence of the new arrangements only at the end of July, and that in my view further reinforces the conclusion that Mr. Kilakos did not understand that the agreement reached with Mr. Heuzé in Singapore had the far-reaching effect he suggested.
174. For a few months after the agreement in Singapore the parties continued to administer the joint venture as they had before. Mr. Heuzé continued to issue deal tickets in respect of parcels sold to MTI and on receipt of a deal ticket Mr. Hawkins would prepare an ITT contract in the established form. However, they soon realised that since all the oil was now being bought by MTI at a pre-determined price and since MTI's requirements were reasonably stable at about 20,000 tons a day, there was really no need for Mr. Heuzé and Mr. Kilakos to continue making regular agreements for in-tank transfers of specific parcels. Accordingly, from about November 1996 the practice of making agreements for the in-tank transfer of specific parcels ceased and Mr. Heuzé stopped issuing deal tickets. Mr. Hawkins simply split each cargo into parcels of about 20,000 tons and issued ITT contracts to MTI in the established form at the rate of about one a day. Broadly speaking, Mr. Hawkins processed the cargoes in the order in which they arrived at Fujairah. The ITT contracts continued to provide for payment by letter of credit a specified period after a deemed stock transfer date against delivery of a stock transfer certificate. MTI continued to open letters of credit under which it paid for the oil and Glencore continued to issue stock transfer certificates following the opening of those letters of credit.
175. As Mr. Crane pointed out, however, under the JV2 arrangements the exchange of information between the parties dwindled almost to nothing. In particular, Mr. Hawkins and Mrs. Gene no longer exchanged inventories and the reconciliation procedure became largely routine. He submitted that that was symptomatic of the fact that Glencore ceased to have any interest in what happened to the oil since it was all to be bought by MTI in any event. Moreover, he submitted, Glencore must have known that MTI was still carrying out substantial amounts of blending since its business depended on its doing so.
176. I cannot accept that view of the matter. In my view the retention of the system of in-tank transfer, albeit in this more streamlined form, is consistent with the conclusion that the parties intended other aspects of their relationship to continue substantially unchanged and that they did not contemplate a fundamental departure of the kind that would be involved in giving MTI complete freedom to dispose of the oil from the moment it was discharged into storage. Various arguments, mainly directed to the financial implications of the arrangement, were canvassed by Mr. Crane with a view to showing that the new arrangement would have been commercially unrealistic if MTI did not have complete freedom to deal with the oil from the outset. These all ultimately rested on the fact that MTI was responsible for all costs associated with the acquisition and storage of the goods and was bearing the entire market risk. It overlooks the fact, however, that it was also MTI that effectively decided what cargoes should be bought for the joint venture and that it was open to MTI to acquire full control over the oil whenever it wished to do so by purchasing it from Glencore. The period of credit established under the JV2 agreement was sufficient, even on the most restrictive view of the ITT contract, to enable MTI to sell products in the bunker market and receive payment before it had to fund the purchase of the oil. It would no doubt have suited MTI better to have had greater freedom to dispose of the oil, and thus to obtain the advantages of improved cashflow, but that was not essential in order to make the arrangements work.
177. In fact, not only the procedure covering sales by in-tank transfer but the manner in which the parties kept their records indicates that neither of them considered that any change of the kind suggested by Mr. Kilakos had been brought about. MTI continued to record oil as entering its own inventory on the ITT date rather than the date of discharge in Fujairah and the parties continued to maintain spreadsheet records on much the same basis as before. Mr. Hawkins opened a new spreadsheet for JV2 while continuing to maintain a parallel spreadsheet for JV1. In the JV1 spreadsheet he entered the relatively small number of cargoes which the parties agreed should be handled under the previous arrangements and also all cargoes which were brought into Fujairah under JV2. At the outset his practice was to transfer those cargoes from the JV1 spreadsheet to the JV2 spreadsheet when an agreement had been made for the sale of the oil, or part of it, to MTI. Later, when the practice of routine in-tank transfers had developed, he transferred cargoes to the JV2 spreadsheet when the time came for them to be sold to MTI and at the same time he recorded the covering in-tank transfers which he continued to enter as withdrawals from inventory on the ITT date. The effect of all this was that at any time Mr. Hawkins' JV1 spreadsheet contained a record of cargoes in store for which no contract of sale had been made and his JV2 spreadsheet contained a record of cargoes in respect of which contracts of sale had been made as well as the dates on which they were treated as leaving the joint venture inventory. This somewhat curious way of recording stock movements is probably best explained by the way in which JV2 had evolved.

Cutterstock

178. The manner in which the parties handled shipments of cutterstock also sheds some light on the nature of their agreement. Cutterstock is oil of very low viscosity which is generally used in small quantities to reduce the viscosity of a blended fuel oil made from heavy components. Mr. Heuzé was well aware that MTI purchased large quantities of Yanbu and other heavy oil specifically for the purpose of blending into 380cst fuel oil, an operation that would generally involve the use of a quantity of cutterstock or similar material. He was aware, therefore, that MTI needed regular supplies of cutterstock.
179. Just two shipments of cutterstock were brought into Fujairah under JV2, a cargo of 96,394 tons on the Hellespont Embassy which discharged in Fujairah at the end of April 1997 and a cargo of 89,631 tons on the Stena Queen which discharged towards the end of July. The bulk of the Hellespont Embassy cargo was sold to MTI under contracts with ITT dates of 6^th, 7^th, 8^th and 9^th October 1997. The final parcel was not sold until the end of December. The bulk of the cargo from the Stena Queen was sold to MTI under contracts with ITT dates ranging from 25^th to 30^th December 1997. In each case Glencore paid storage charges for the parcels sold to MTI for the period up to the end of the month preceding the ITT date.
180. Mr. Crane submitted that it would make no commercial sense for MTI to hold such large quantities of cutterstock in store for such a long time without being able to make any use of it and that Glencore was well aware that it was being used in small quantities throughout the intervening period even though at that stage it had not been formally sold to MTI. There is no doubt that there was an unusual delay in issuing ITT contracts in respect of the Hellespont Embassy cargo. Part of that was due to the need to await pricing information under the contract with the prime supplier, as was envisaged when the arrangement to supply this cargo was originally made, but not all of it can be put down to that because the relevant pricing period ended on 30^th June.
181. A special rate of $2.50 per ton for the first month and $0.50 per ton per month thereafter had been agreed for the storage of cutterstock back in September 1995. During the summer and autumn of 1997 MTI rendered storage invoices to Glencore on the basis that the whole of the cutterstock delivered by the Hellespont Embassy and the Stena Queen remained in store. From November the invoices reflect reductions in the amount held in store in accordance with the in-tank transfers issued in October. Mr. Kilakos said that Mr. Heuzé had not only agreed but had suggested that formal sales of the cutterstock to MTI should be deferred as a way of giving MTI an extended period of credit, but that the oil itself could be used by MTI before any ITT contracts had been made.
182. I do not think that there is much doubt that the cutterstock from the Hellespont Embassy and Stena Queen was used, as Mr. Kilakos said, "slowly, slowly" during the summer and autumn of 1997. Nor is there any doubt that there was agreement that the pricing of the Hellespont Embassy cargo would have to be deferred because of the terms of the contract with the prime supplier. Mr. Crane submitted that Mr. Hawkins' explanation that part of the delay was due to an oversight on his part did not ring true and tended to support Mr. Kilakos' evidence that there had been an agreement with Mr. Heuzé of the kind he described. I find it difficult to accept, however, that Mr. Heuzé agreed that MTI could use the oil before there had even been an ITT contract. The pricing period for the main parcel of cutterstock on the Hellespont Embassy was April-June and for the balance September-October. The first period ended just over two months after the vessel's arrival at Fujairah and an ITT contract could have been issued at that time. At that point both parties would have treated the oil as delivered to MTI and, if necessary, a letter of credit could have been opened. There is no reason to think that MTI had such a pressing need for new supplies of cutterstock that it could not wait that long to begin using this particular cargo. It would have been a very unusual thing for Mr. Heuzé to agree that MTI should be able to make use of oil before formalities of any kind had been put in place, but if he had done so, it is difficult to see why MTI should have continued to charge, or Glencore should have paid, storage fees as if none of the cargo had been used, or why MTI's internal records should reflect transfers only on the relevant ITT dates. The evidence of the way in which the parties dealt with cutterstock during the period of JV1 indicates that there would have been no real difficulty in identifying the quantity used each month so as to enable storage invoices to reflect consumption. I am satisfied that MTI was indeed making use of the cutterstock prior to the issue of any ITT contracts, but I can see no reason to conclude that either Mr. Heuzé or Mr. Hawkins was aware of the fact.

Glencore's purchases from the joint venture

183. In seeking to rebut Glencore's case that MTI was not permitted to dispose of oil in any way until a letter of credit had been opened in respect of it in accordance with the terms of the ITT contracts Mr. Crane relied on a number of instances in which Glencore itself bought oil from the joint venture. The cargoes in question were shipped out of Fujairah on the vessels Shibumi, Shoko, Erissos, Knock Buie, Addax and Amyndas. In each of these cases the oil in question was sold by Glencore to MTI and then re-sold by MTI to Glencore on terms f.o.b. Fujairah. In each case the sale by Glencore to MTI was carried out using the established form of ITT contract which described the oil in question as being held on board the Metrotank. Letters of credit were opened by MTI in respect of the price of the oil and Glencore issued stock transfer certificates in the same way as it did in respect of other in-tank transfers. However, in none of these cases was the relevant cargo discharged into storage and for this reason alone an in-tank transfer was not an appropriate form of contract. Mr. Crane submitted that these transactions demonstrate, contrary to Glencore's case, that the ITT contracts were used for convenience or simply out of habit in circumstances where the parties were both well aware that they did not reflect their true intentions and in some cases could not be performed in accordance with their terms. Mr. Schaff, on the other hand, submitted that these were exceptional cases in which there had been specific agreement between Mr. Heuzé and Mr. Kilakos on each occasion as to how the contract should be performed. They could not be treated as evidence of a general acceptance on the part of Glencore that MTI should be free to ignore the terms of the ITT contracts by disposing of oil as and when it was commercially desirable to do so. In order to determine this question is necessary to look at the individual circumstances of each case.

(i) Erissos (b/l 16.6.96)

184. Glencore agreed to buy the cargo on 3^rd June 1996, about ten days before the discussions between Mr. Heuzé and Mr. Kilakos in Singapore. However, it was subsequently agreed that the cargo would be sold on by MTI to buyers in Singapore under the JV2 arrangements. The cargo was not discharged at Fujairah and Mr. Hawkins noted that fact in his JV2 spreadsheet. Despite that, Glencore issued ITT contracts as if the cargo had been discharged and was to be transferred to MTI by in-tank transfer. Mr. Crane submitted that the issue of ITT contracts in respect of a cargo which was known not to be in store demonstrates the fact that Glencore treated these contracts as little more than a formality and was unconcerned about the physical location of the oil which it had agreed to sell to MTI. Mr. Heuzé accepted that the cargo should not have been sold under ITT contracts and agreed that in this case Glencore had no security for the price until MTI opened its letter of credit. However, I do not think that this transaction can be regarded as typical. The evidence of both Mr. Heuzé and Mr. Kilakos makes it clear that this was an unusual situation. The cargo appears originally to have been destined for storage, and it may be that Mr. Hawkins prepared ITT contracts in that expectation. However, after tests had been carried out on the cargo at Fujairah Mr. Heuzé and Mr. Kilakos agreed that it should not be discharged but sold on. Obviously the documents issued to cover the sale by Glencore to MTI were entirely inappropriate, but I do not think that one can draw from that the inference that in general Glencore regarded the ITT contracts as nothing more than an administrative convenience.

(ii) Shoko (b/l 30.1.97)

185. On 14^th January 1997 Glencore agreed to buy from MTI 250,000-300,000 barrels of LRM and 3,000-6,000 tons of Kangan condensate f.o.b. Fujairah. Shortly afterwards Glencore had agreed to buy for the benefit of the JV2 arrangement a cargo of about 50,000 tons of LRM under a back-to-back contract with MTI which in due course was shipped to Fujairah on the vessel Shoko. On 24^th January 1997 Glencore asked MTI to discharge part of the Shoko's cargo but to keep the rest on board and blend it with other oil including the parcel of Kangan condensate to which I have just referred. The operation was completed by 31^st January. Glencore paid MTI for the cargoes of LRM and Kangan condensate on or about 28^th February 1997, but Mr. Hawkins did not issue ITT contracts to MTI in respect of them until 27^th February and MTI did not open a letter of credit until 2^nd April. A stock transfer certificate was issued on 3^rd April. Here again, Glencore had paid both to obtain the oil for the joint venture and to purchase the oil from MTI without obtaining any security for the performance of MTI's obligation.

(iii) Knock Buie(b/l 7.2.97)

186. In January 1997 Glencore agreed to buy 53,585 tons of LRM from Saudi Aramco as prime supplier. The cargo was shipped on board the vessel Cherry at Jubail early in February 1997. On 18^th February Mr. Hawkins instructed MTI to tranship 26,000 tons of the cargo from the Cherry to the Knock Buie on arrival at Fujairah rather than discharge it into store. In the event the Knock Buie was late reaching Fujairah and the instructions were changed to allow discharge into the Stresa to await the arrival of the Knock Buie. The contract for the sale of 26,000 tons of LRM by MTI to Glencore f.o.b. Fujairah was not issued until 26^th February, with title passing at port of loading.
187. Despite the fact that half of it had been transhipped onto the Knock Buie, Capt. Margaritis, who was responsible for operational matters in MTI's Athens office, sent Glencore a telex stating that the whole of the Cherry's cargo had been discharged into storage, and indeed the parties treated that as being the case. Mr.Hawkins said that he had not noticed the error in Capt. Margaritis' telex, but in the circumstances I find that surprising. In any event, three ITT contracts covering the whole of the cargo on the Cherry were issued by Mr. Hawkins with ITT dates of 4^th, 5^th and 6^th March 1997 and these parcels appear in his spreadsheet as leaving the joint venture inventory on that date. The covering letter of credit was not opened until 4^th April 1997 and it follows that in the intervening period Glencore was not secured for the purchase price due from MTI. Moreover, if the ITT contracts were taken at face value, MTI had no title to sell the oil at the time when it was shipped on the Knock Buie, or indeed at any time before the stock transfer certificates were issued a few days after the letter of credit had been opened.

(iv) Shibumi (b/l 9/11/97)

188. On 30^th October 1997 Glencore agreed to buy from MTI a cargo of about 80,000 tons of 280cst fuel oil f.o.b. Fujairah. The vessel Shibumi was nominated to load the cargo which was shipped on 9^th November. Glencore had asked that the cargo should be made up of a blend of oil from three other vessels, the Emerald Star, Arma and Mint Prosperity to ensure that the cargo met the particular requirements of Glencore's customer. The Emerald Star and Mint Prosperity, each of which was carrying a joint venture cargo, discharged direct into the Shibumi, in the case of the Emerald Star only about 3,000 tons, but in the case of the Mint Prosperity the whole cargo of about 78,000 tons. Glencore paid for the Shibumi cargo on or about 28^th November, even though MTI had not by that stage paid for the cargoes on the Emerald Star or Mint Prosperity. In fact Mr. Hawkins did not issue ITT contracts in respect of the cargoes on those vessels until 2^nd February and 15^th December respectively. It follows, as Mr. Heuzé accepted, that although by the end of November Glencore had paid both the prime supplier (or MTI under a back-to-back contract) and MTI for the goods, it had no security for the amount owed by MTI in respect of the purchase price due from it.

(v) Addax and Amyndas

189. This was a transaction of a similar kind, although in the event the collapse of MTI intervened before it could be completed. On 8^th January 1998 Glencore agreed to buy from MTI on back-to-back terms a parcel of 50-54,000 tons of LRM f.o.b. Jubail. The vessel Addax was nominated to ship the parcel. On 28^th January, as recorded in his deal ticket of that date, Mr. Heuzé agreed with Mr. Kilakos to buy 40-45,000 tons of LRM to be transhipped onto the vessel Amyndas from the Addax at Fujairah. Mr. Heuzé recognised that if this transaction had been completed it would have involved a sale by Glencore to MTI by in-tank transfer in the established manner and the purchase by Glencore from MTI of part of a parcel which it had only just bought from MTI under the back-to-back contract. In other words, ITT contracts would have been issued in respect of a parcel which had never been discharged into storage in Fujairah. So this is another example of Glencore's purchasing and taking delivery of a cargo from MTI at a time when, if Glencore's case is correct, MTI had not acquired title to it.
190. These five examples extend over the whole period of JV2. Mr. Crane submitted that they demonstrate a willingness on the part of Glencore to buy, and to take delivery from MTI of, oil to which, under the strict terms of the ITT contracts, MTI had not acquired title and to do so in circumstances where it retained no security for the payment by MTI of the price due from it. He submitted that that supported the conclusion that Glencore was generally content for MTI to dispose of oil even before any ITT contract had been entered into and certainly before a letter of credit had been opened in respect of the price due. This too, he submitted, was consistent with the parties' treating the ITT contracts as no more than a convenient mechanism for accounting for the price due in respect of oil which it had already disposed of.
191. These transactions are certainly anomalous in their use of the standard form of ITT contract to regulate the sale of the oil by Glencore to MTI and the consequent recording of the movement of oil as if it had not left the inventory until the relevant ITT date. They are also a striking example of transactions under which Glencore did not obtain security for the payment by MTI of the price of the goods. As such they might be said to reinforce to some extent the conclusion that Glencore was willing to allow MTI to dispose of oil before a letter of credit had been put in place. Do they, however, indicate that it was generally willing to allow MTI to dispose of oil before any contract of sale of any kind had been put in place and that the parties had reached the point at which that had become the agreed basis of their relationship? I do not think so. These transactions have to be viewed in their context, three aspects of which are of particular significance. The first is that under the arrangements as modified in June 1996 MTI, as the party with the obligation to purchase and dispose of all the oil brought into the joint venture, controlled the purchase of oil for that purpose and organised it to meet its own requirements. Although Glencore became the owner of oil bought for the joint venture, Mr. Heuzé and Mr. Silverberg recognised that it could not simply make use of these cargoes for its own purposes without the agreement of MTI and without giving MTI the benefit of any profit which might be derived from a sale at market price and (as they seemed to think) the $0.75 mark-up which it would obtain under the joint venture agreement. I accept Mr. Silverberg's explanation that the rather cumbersome process of sale and re-purchase demonstrated in these transactions was seen as a way of preserving the essence of the joint venture relationship.
192. The second factor is the number of transactions which the parties entered into during the period of JV2. It was agreed that the first cargo to be handled under new arrangements should be a cargo of Yanbu carried on the vessel Aladdin under a bill of lading dated 18^th June 1996. Between June 1996 and February 1998 when the venture came to an end over 180 cargoes of oil were bought and sold under the JV2 arrangements. In my view the transactions on which Mr. Crane relied represent too small a proportion of the whole to be significant. The third factor is the way in which in-tank transfers were administered. From the latter part of 1996 ITT contracts were issued by Mr. Hawkins in the manner mentioned earlier without any specific agreement between Mr. Heuzé and Mr. Kilakos. These five transactions, on the other hand, were not typical in that each of them resulted from a specific agreement between Mr. Heuzé and Mr. Kilakos. I think it likely that, having been told that Glencore had agreed to buy a cargo which was being brought into Fujairah under JV2, Mr. Hawkins simply continued to operate the established ITT procedure without giving any particular thought to the time at which title would pass from Glencore to MTI and back again, or, since the cargo itself was remaining in the possession of Glencore, to the question of security for the payment by MTI of the price. Mr. Hawkins dealt with a very large number of transactions every day. He said that he did not always read telexes he received with great care and although I find it difficult to accept that he was unaware of some of the more practical aspects of these transactions, I find it equally difficult to accept that he gave any real consideration to their legal implications.

The effect of storage in common bulk

193. Having reached certain conclusions as to the parties' intentions with regard to the storage and handling of cargoes it is possible to consider Mr. Crane's submission that the effect of the joint venture agreements was to entitle MTI to dispose of oil held at Fujairah as from the moment at which it was discharged into storage.
194. Mr. Crane submitted that, both under the law of Fujairah and under the law of England, where goods of a similar kind are commingled in store the owners of the contributing parcels become owners in common only of the physical bulk to which their goods have contributed. They do not acquire any proprietary interest in other goods of the same kind held elsewhere in the same store, even if the storekeeper is under an obligation to redeliver on demand an equivalent quantity of goods of the same kind from his general stock. He submitted that the contract between the parties in the present case was of the latter kind and that the relationship between the parties was therefore one which the law of Fujairah would characterise as qardh. The legal incidents of qardh are discussed at length in my judgment on Phase 1.
195. In relation to the position under English law Mr. Crane reminded me of the decisions in South Australian v Randell (1869) L.R. 3 P.C. 101 and Chapman v Verco (1933) 49 C.L.R. 306 , both of which I considered in some detail in my judgment in Phase 1. However, I am not concerned in the present case with the effect of English law, and although I do not think that the law of England and the law of Fujairah differ significantly on this question, I do not propose to say any more about these cases here.
196. In my judgment in Phase 1 I discussed at some length the effect of Articles 710, 975 and 992 of the Commercial Code of the UAE and the principles underlying them. The essential characteristic of qardh is an intention to permit the person to whom the goods are delivered to dispose of them as his own, even though that involves their consumption or destruction. This appears most clearly from Article 992. Since a transaction of this kind is intended to give the recipient complete dominion over the goods, it is inconsistent with title remaining in the depositor. Title therefore passes to the recipient. This remains the case even though the recipient is under an obligation to redeliver a similar quantity of goods of a similar kind to the depositor on demand. He may even redeliver the very same goods, but the effect of doing so is merely to transfer property in them back to the original owner. The existence of a relationship of qardh, therefore, ultimately depends on the intention of the parties, in particular an intention on the part of the depositor that the recipient should be entitled to dispose of the goods as his own. Ascertaining the intention of the parties is therefore of great importance, both for the reason just mentioned and because in transactions of this kind the law of Fujairah generally seeks to give effect to the intention of the parties.
197. It is necessary in this context to distinguish two quite separate questions. The first relates to the intention of the depositor with regard to the purpose of the deposit. The second relates to the intention of the parties to the storage contract with regard to the manner of storage, the degree of segregation, if any, of the goods in store and the identity of the bulk to which those goods contribute if commingling in storage is allowed. As far as the first is concerned, I am satisfied, for reasons I have already given, that at no stage in the parties' relationship did Glencore itself intend property in oil delivered into the floating storage to pass to MTI immediately upon delivery. That would not be determinative of the question, however, if one could see from the circumstances in which delivery occurred that MTI was entitled to make use of it immediately in a way which involved its consumption in one form or another. If that were the case it would broadly equate to the situation in South Australian v Randell and would bring the deposit within the scope of Article 992 of the Civil Code. If, therefore, Glencore had intended MTI to be free to use the oil for blending at its discretion from the moment of its delivery into store, the transaction would have been one of qardh and property in the oil would have passed to it at once. However, for reasons I have already given I am satisfied that that was not the case.
198. The second question relates entirely to the manner of storage contemplated by the parties. It was inherent in the concept of the floating storage facility as operated by MTI that there was a degree of flexibility in capacity by virtue of the fact that the number of vessels comprising the facility could be increased or reduced to meet fluctuation in demand. The capacity of the Metrotank was substantial, but it had been adapted to act as a blending terminal and could not provide sufficient tankage to cover all MTI's needs. Moreover, it was not fitted with a double valve segregation system, so its ability to protect the integrity of cargoes was limited. From the outset, therefore, other vessels were chartered to increase the capacity of the floating storage facility or, as in the case of the Mount Athos, to provide complete segregation of a particular cargo or grade where that was required. Glencore was well aware of the manner in which the facility was operated and it is clear that although the documents on both sides often refer to the storage facility as 'the Metrotank', both parties understood that to be shorthand for the facility as a whole.
199. Apart from the storage of Iranian straight run fuel oil on the Mount Athos, I think there is little doubt that both Glencore and MTI treated the whole of the storage facility operated by MTI a single unit. Although MTI was bound to keep oil segregated by grade, it was given a free hand to store oil brought into the store in commingled bulk in whatever way it found most convenient. There was nothing in the agreements between the parties which restricted MTI's operational discretion in relation to the location of oil stored for the benefit of the joint venture, nor was any record kept of the tanks into which incoming cargoes were discharged or subsequently moved. Mr. Crane accepted that that was so, but he did not accept that the parties had thereby agreed to treat the whole of the tankage containing any particular grade as constituting a single bulk. He submitted that since it was impossible to identify the physical bulk to which Glencore's oil had contributed the intention was that MTI should redeliver oil from its general stock when called upon to do so. In those circumstances no bailment of any kind continued to exist and MTI became owner of the oil.
200. I am unable to accept that argument because it seems to me to reflect neither the parties' intention nor the commercial realities. The fact that the parties treated the transaction as involving the storage of oil in the first instance and the storage facility as a single unit points strongly to the conclusion that the whole of MTI's stock of any given grade in the floating storage was to be regarded as between the two of them as a single bulk. Whatever contrary inference might otherwise have arisen from the provision in the May 1994 storage agreement that holding certificates should be given by the masters of vessels on which oil was stored disappears entirely given the fact that it was immediately allowed to lapse without protest from either party. There is nothing of which I am aware in the law of Fujairah which prevents parties from doing business on that basis. Indeed, when dealing with a transaction of this kind the law of Fujairah gives effect to the parties' intention so far as it is possible to do so. Commingling does not involve consumption or use of a kind which would otherwise turn a relationship of bailment into one of qardh, even though it involves a loss of specific identity. The only real question in a case such as the present is whether the person with whom the goods have been deposited has agreed with the depositor that the whole of the goods of a particular kind which he holds in stock shall be treated as part of a single bulk for these purposes. That is a matter to be decided on the facts of the individual case. In the present case I think that from May 1994 onwards the parties conducted the business of the joint venture on that basis that all stocks of oil of any particular grade held in the floating storage constituted a single bulk.
201. Mr. Crane submitted, however, that that is not sufficient to determine Glencore's proprietary interest in the bulk where other depositors have also contributed to it, possibly on different terms and he drew my attention to the position papers served by Caltex, Mobil and Arexco in this case, each of which expressly or by implication asserts that MTI was not entitled to commingle oil held for it in store with oil owned by MTI itself or third parties. I can certainly see that where there are multiple depositors, some of whom have deposited oil on terms that it shall be kept in segregated storage, there may be difficulties in ascertaining proprietary interests if there has been unauthorised commingling or a failure to keep records of a kind which enable the oil delivered into store by any particular depositor to be identified. However, issues of that kind can only be decided after hearing evidence and submissions from all interested parties. They do not in my view affect the essential principles applicable to the relationship between Glencore and MTI, but in any event since the other oil claimants are not represented at this trial they must be reserved for a later occasion.
202. Whatever may have been the position under JV1, however, Mr. Crane submitted that the position was clearer under JV2 inasmuch as it was an essential element of the modified arrangements that MTI would purchase from Glencore all the oil delivered into storage. Accordingly, he submitted that the situation which existed under JV2 is similar to that considered in a number of English cases where components or raw materials have been delivered to a manufacturer under a contract containing a reservation of title clause. He submitted that in such a case the mere presence in the contract of a clause reserving title to the seller pending payment for the goods does not prevent the buyer from making use of the goods in the manufacturing process, nor, he submitted, does it give rise to an initial relationship of bailment between the buyer and seller.
203. In Benjamin's Sale of Goods, 5^th ed., §5-139, to which Mr. Crane drew my attention, it is suggested that where goods are sold to a manufacturing or trading company under a contract containing a reservation of title clause, particularly where a period of credit is allowed, it can scarcely be supposed that the buyer is meanwhile to have no right to consume or resell the goods in the ordinary course of business and that a term may be implied to that effect to give business efficacy to the contract. I would not dissent from that as a general proposition, provided it is borne in mind that whether, and if so, under what circumstances, the buyer is entitled to make use of the goods before paying for them is essentially a matter for agreement between the parties. No doubt in a case of that kind it will usually be clear from the nature of the contract and the context in which it is made that the buyer is to have the right to make use of the goods in the ordinary course of business, but it still remains open to the parties to agree otherwise. The relationship between the parties in the present case, even as modified in June 1996, was not a simple case of a sale of goods by a supplier to a trading company or manufacturer. It had its origin in a complex relationship involving the storage and sale of oil under a joint venture which included sales to third parties as well as to MTI. Glencore had not agreed prior to June 1996 that MTI should be entitled to make use of oil delivered to it for its own purposes prior to an agreement for sale by in-tank transfer and despite the modification to their relationship brought about in June 1996, the parties continued to treat the arrangement as a joint venture under which oil was held in storage for Glencore pending sale to MTI. Against that background it is in my view impossible to treat the discharge of oil into the possession of MTI at Fujairah as nothing more than the delivery of goods under a contract of sale and to imply into the over-arching agreement constituted by JV2 a term that MTI was entitled to dispose of it freely as from that moment.

Contango

204. It was suggested that some light could be shed on the nature of the parties' relationship under the JV2 arrangements by the circumstances surrounding a contango play which Glencore sought to make for the benefit of the joint venture at the beginning of 1998. A commodity market is said to be in 'contango' when the price of goods for prompt delivery is significantly lower than the forward price. If the difference is sufficiently great a trader who is able to buy goods and hold them for delivery in the future can hope to make a profit on the transaction, provided, of course that the market does not decline in the meantime. A profit can be assured, however, if the transaction is hedged in the futures market. To do that it is necessary for the trader to sell a corresponding quantity of goods forward in the paper futures market at a fixed forward price and also to enter into a matching purchase in the paper futures market at the price ruling at the date of delivery in order to balance his book. The difference between the purchase price of the physical oil on the spot market plus the cost of storage and insurance for the period in question on the one hand and the sale price under the paper futures contract on the other represents a 'locked in' profit. When the time for performance of the futures contracts arrives the oil in store is sold on the spot physical market. If the market is high enough for the sale price to exceed the cost of purchase, the trader will take a profit on the physical transactions, but the loss on the paper transactions will limit that profit to the 'locked in' amount. If the market is not high enough, the trader will make a loss on the physical transactions but will take the 'locked in' profit under the paper transactions.
205. In early 1998 the fuel oil market was in deep contango in that the price of oil for immediate delivery was markedly lower than the price on the futures market for the period February-April 1998. In order to take advantage of that situation Mr. Heuzé suggested to Mr. Kilakos that the joint venture should buy a number of cargoes for the specific purpose of making a contango play. They agreed that Glencore would buy the cargoes from prime suppliers for the benefit of the joint venture and undertake the hedging transactions. MTI would be paid a fee of $1.95 per ton per month for storing the cargoes and would buy in the cargoes from the joint venture at market prices over a period corresponding to the range of dates by reference to which the paper transactions were priced. In the event six cargoes were bought for this purpose.
206. In a normal contango play the trader will sell the oil in the open market and must have the oil available for delivery if he is to take the locked in profit rather than simply speculate by making a forward short sale. The present case was unusual in that the buyer was identified from the outset as MTI which was also storing the oil and it was common ground that in the event MTI had not kept the oil in store but had disposed of it in advance of the pricing dates. In his witness statement Mr. Heuzé said that he was surprised to learn that MTI had disposed of these cargoes given the fact that they had been bought specifically for the purposes of the contango play. Mr. Crane submitted that at this point in his evidence Mr. Heuzé was being somewhat disingenuous in that he was seeking to emphasise the need for MTI to retain possession of the oil primarily in order to reinforce Glencore's case in relation to the general nature of the joint venture arrangement. He submitted that since MTI had obligated itself to buy the oil at the future market price in any event, it was entirely sensible for it to reduce the risk of loss by selling some of it to third parties when a suitable opportunity presented itself.
207. I do not think that is really a fair criticism. The ability to deliver physical goods is an essential element of a contango play and one which had been explained to Mr. Kilakos. MTI was to receive an enhanced fee for storing the oil in question. Technically Glencore was obliged to deliver oil to MTI on the forward date and no agreement had been made giving MTI the right to dispose of it in the meantime. No contracts of sale had been made between them, although it was clearly contemplated that in-tank transfers would be effected in due course. In these circumstances I think Mr. Heuzé was entitled to be surprised when he discovered that much of the oil had already been disposed of. Mr. Crane also submitted that the way in which the parties discussed storage arrangements for this contango play indicated that in general under JV2 they did not expect cargoes to remain in storage for long, if at all. I cannot accept that. In my view nothing in the discussions about the contango play sheds any real light on the general nature of the relationship under JV2. In the end I think this whole topic is one of marginal relevance to the issues I have to decide.

The collapse of MTI

208. During the last few months of 1997 there was a prolonged and very substantial fall in the market price of fuel oil to the point where oil which MTI had agreed to buy at prices equivalent to about $100 per ton could be sold at only $50 per ton or thereabouts. That inevitably placed MTI under great financial pressure. However, MTI had a continuing requirement for oil in order to supply its customers and Glencore continued to issue ITT contracts at the rate of about 20,000 tons per day so that the quantity transferred each month during November and December 1997 and January 1998 remained fairly constant.
209. The fall in the fuel oil market meant that MTI could no longer sell bunkers at prices which fully reflected the cost of the components. That inevitably placed MTI under considerable financial pressure, but I do not think that Glencore was conscious of there being any real risk that it would be unable to meet its commitments until towards the end of January 1998. When Mr. Kilakos met Mr. Heuzé in Athens on 7^th January he admitted that MTI had suffered financially, but said that it could weather the storm, and Mr. Heuzé received similar reassurances from Mr. Griffin. At that stage, therefore, there did not seem to be any reason for alarm.
210. On 30^th January Mr. Kilakos and Mrs. Gene went to Paris to see MTI's main bankers, Banque Trad-Crédit Lyonnais ("BTCL"). The meeting came at a critical time because MTI had reached the limit of its credit line and the bank had decided not to open any further letters of credit until after the meeting with Mr. Kilakos. In the event the bank did not open any further letters of credit and MTI made the next payment to Glencore in cash. Subsequent events, in particular MTI's application to the Greek courts on 27^th February for a suspension of payments with effect from 1^st February 1998, leave me in no doubt that on 30^th January BTCL indicated its refusal to extend any more credit to MTI.
211. At the beginning of the following week both Mr. Heuzé and Mr. Palacios travelled to Paris. Mr. Palacios went in order introduce Mr. Kilakos to certain banks and to make a presentation in the hope of obtaining finance for a joint shipping venture which Glencore and MTI had in mind. Mr. Heuzé had been told by Mr. Kilakos that he would be in Paris that week and went to meet him to discuss various matters. They all had dinner together on the evening of Monday 2^nd February, but nothing was said on that occasion about MTI's difficulties.
212. The picture became clearer the next day. According to Mr. Heuzé, when he and Mr. Palacios met Mr. Kilakos that morning Mr. Kilakos told them that MTI had serious cash flow difficulties and could open no more letters of credit. Mr. Palacios' recollection was rather different. He said that Mr. Heuzé had mentioned to him that morning that he had heard that Mr. Kilakos had had a difficult meeting with BTCL on the Friday, but that he had not said anything to make him unduly concerned. He had visited two banks during the morning in company with Mr. Kilakos and his son, John. Later that day Mr. Kilakos had been a little more forthcoming about MTI's difficulties, but did not give any precise details. Later Mr. Heuzé, who by then was on his way to Tehran, telephoned him from the airport. By that stage, Mr. Palacios said, Mr. Heuzé had become more concerned about MTI's position. He recalled his being concerned about whether MTI had hedged its fuel oil position, but he did not recall his saying anything about its inability to finance any more letters of credit.
213. If Mr. Kilakos had made it clear to Mr. Heuzé first thing on the Tuesday morning in the presence of Mr. Palacios that MTI could no longer open letters of credit, I am sure that Mr. Palacios would have remembered that and would not have embarked on meetings with the banks as if nothing had happened. I think it likely, therefore, that the conversation to which Mr. Heuzé referred took place in circumstances different from those he described. However, that may be of little significance since Mr. Kilakos accepted that he had told Mr. Heuzé before he left Paris that MTI would not be able to obtain further financing from its banks for the time being. Later in the day Mr. Kilakos told Mr. Palacios that MTI was in some financial difficulty, although he did not give him the impression that the problem was a serious one.
214. After Mr. Kilakos had told him of MTI's financial problems Mr. Heuzé telephoned the London office and spoke to Mr. Silverberg. They decided that it would be wise to send someone out to Fujairah to report on the state of the inventory, although according to Mr. Heuzé they did not at that time think that oil was missing. Mr. Heuzé left Paris for Tehran during the afternoon of 3^rd February. He returned to London on Friday 6^th February via Frankfurt. Mr. Kilakos left Paris on the evening of Wednesday 4^th February arriving in Fujairah the next morning.
215. As might be expected, there was a flurry of communications during the latter part of the week. The sequence and details of these communications is clouded by the fact that all the main participants were on the move at one time or another. There was a good deal of argument about how much information was already available to Glencore by the time Mr. Heuzé spoke to Mr. Silverberg and what steps were taken to obtain information from other sources, especially MTI. That was directed mainly to showing that Glencore, through Mr. Heuzé and Mr. Hawkins, had long been aware that MTI was not holding in store oil in respect of which ITT contracts had been issued but in respect of which letters of credit were still to be opened.
216. Mr. Heuzé said that when Mr. Kilakos told him that MTI was unable to open any further letters of credit he immediately asked him where the oil was which MTI was holding in store, to which Mr. Kilakos replied that it was "on the ships". Mr. Kilakos, however, denied that any exchange of that kind had taken place. Mr. Crane submitted that that would be an odd question for Mr. Heuzé to ask if, as he said was the case, he understood that MTI was holding in storage all the oil in respect of which no letter of credit had yet been opened. I do not think it is such an odd question, however. When a depositor discovers that a company holding his goods in store is in serious financial difficulty, one of the first things likely to cross his mind is that his goods may have been misappropriated. He is quite likely, therefore, to seek confirmation that his goods are physically safe and still in store. In the present case the goods were being stored in ships, so I do not find it surprising or of particular significance that Glencore should wish to obtain more precise information about the location of specific parcels. The fact that as early as 3^rd February Glencore thought that it might be desirable to send someone out to Fujairah to report on the inventory does, as Mr. Crane submitted, sit uneasily with Mr. Heuzé's evidence that he did not at that stage think that there was a physical loss, but in fact Mr. Garrett who was chosen for the job did not leave for Fujairah until 8^th February, almost a week later. That hardly indicates that Glencore thought there was an immediate problem of any magnitude.
217. In his witness statement Mr. Heuzé said that Mr. Silverberg had asked him in the course of their conversation on 3^rd February to obtain from Mr. Kilakos details of the individual capacities of the storage vessels. I do find that a little more surprising because the natural assumption would be that there was sufficient capacity to accommodate whatever quantity ought to have been in store. It does tend to suggest that Glencore knew that in the past some oil had or might have been disposed of before letters of credit had been opened, but not how much. Beyond that, however, I am not sure that it takes the matter very far.
218. In his statement Mr. Silverberg said that the information he had requested from Mr. Kilakos had been received from MTI the next day (Thursday) and that when it had been analysed internally it had become apparent that the storage capacity available to MTI was about one million tons less than was required to accommodate all the oil it should have been holding for Glencore. After a meeting had been held to discuss the problem he spoke by telephone to Mr. Kilakos and arranged to meet him in Athens on Saturday 7^th February to discuss the position generally. The precise nature of the information said to have been received from MTI and the manner in which it had been supplied remained somewhat obscure, however. The impression one gets from Mr. Silverberg's statement is that it was provided in written form, probably by fax, but no fax of that kind appears to have been sent by MTI and in cross-examination Mr. Silverberg said that the information had been given to him by Mr. Kilakos in a previous telephone conversation earlier the same day after he had contacted him for that purpose. Mr. Silverberg's evidence in this respect is at best confused and I do not feel able to place any great weight on it. Some written details of the storage capacity available to MTI were provided by Mr. Kilakos, but that was not until the meeting which took place in Athens on Saturday 7^th February. Mr. Kilakos denied having spoken to Mr. Silverberg that week before they met in Athens; his recollection was that it was Mr. Heuzé who had asked him to attend a meeting with senior representatives from Glencore. I prefer his evidence on this question.
219. As soon as MTI's problems came to light Glencore turned to its solicitors, Clyde & Co, for advice. On 6^th February in preparation for the meeting with Mr. Kilakos the next day they prepared a draft of an agreement to which I shall return in more detail a little later. Clause 1 of that draft recorded that Glencore had delivered about 2.8 million tons of oil to companies in the Kilakos group of which about 1 million tons could not be accounted for. These figures were obviously supplied to Clyde & Co. by Glencore. How did Glencore arrive at the figure of 1 million tons?
220. On Thursday 5^th February Mr. Hawkins sent a fax to Mr. Kilakos attaching a copy of his JV1 spreadsheets for 1997 and 1998 which showed cargoes held in store awaiting transfer to MTI. He concluded by asking Mr. Kilakos to advise the quantities and locations of the oil currently being held to the order of Glencore. This was an unusual message in several respects. In the first place, Mr. Hawkins did not normally correspond with Mr. Kilakos about anything; his regular correspondent was Mrs. Gene, particularly about inventories which were essentially a routine operational matter with which Mr. Kilakos would not be concerned. Secondly, the message was sent to Mr. Kilakos' home as well as the office in Fujairah. This suggests that it was regarded as a matter of particular importance. Thirdly, this is the only example of a request being made by Glencore for information about the location in which oil was held. This all suggests that Glencore was seeking as a matter of some urgency information which it did not already have in its possession.
221. In the evening of 5^th February Mr. Silverberg travelled to Israel leaving Mr. Dreyfus in charge of the situation in London. Mr. Heuzé did not return to London from Tehran until the morning of Friday 6^th February. The information needed to produce the draft agreement was provided to Clyde & Co by Mr. Palacios on Friday 6^th February. He could not remember what had been the source of the figure for oil which could not be accounted for: he thought that he had probably taken a round figure based on the various discussions held within Glencore the previous day. Mr. Cooke submitted that the figure of 1 million tons in fact represented in round terms the amount of oil which MTI had drawn from joint venture stock without any covering ITT contract. In other words, it reflected and further supported the insurers' 'third way' case. He submitted that the figure could not have been calculated by Glencore and so must have been provided by MTI itself.
222. It is quite true that MTI's records show that as at 6^th February it had drawn 964,031 tons of oil which had not been the subject of any ITT contract. It is also true that the figure in MTI's records is quite close to 1 million tons. However, with the benefit of hindsight it can be seen that the figure of 1.1-1.2 million tons which Mr. Silverberg said Glencore had calculated as missing from information available to it on 5^th February was quite close to the figure of 1.374 million tons calculated by Mr. Silverberg during the meeting on 7^th February and must have included oil which had been the subject of ITT contracts as well as oil which had not. I do not think, therefore, that it is a figure which can have come direct from MTI since Mr. Kilakos did not regard oil which had been sold to MTI as still held to Glencore's order after the ITT date. Moreover, if the figure had been supplied in the way Mr. Cooke suggested, I should have expected Mr. Palacios to have remembered that. Mr. Crane submitted that Glencore through Mr. Heuzé had kept itself informed of the storage capacity maintained by MTI at Fujairah and had known for some time that MTI could not be holding in store all the oil still awaiting sale to MTI as well as oil which had been sold but for which letters of credit had yet to be opened. He accused Glencore of "hiding behind its document trail" as soon as it became apparent that MTI was no longer able to open any more letters of credit.
223. The suggestion that Mr. Heuzé, Mr. Silverberg, Mr. Palacios and other employees of Glencore were all involved in an elaborate attempt during the first week in February 1998 to suppress the extent of their knowledge about MTI's operations is one which I am unable to accept. For the reasons I have already given I accept that Mr. Heuzé had a certain amount of information about the vessels available to MTI at Fujairah; he did not pretend otherwise. I do not accept, however, that he was conscious of the fact that the total storage capacity maintained by MTI was significantly less than would have been necessary to contain all the oil awaiting letter of credit. Although the figure for the shortfall in oil calculated by Mr. Silverberg on 5^th February is reasonably close to the difference between the quantity of oil which Glencore said MTI should be holding to its order (2.832 million tons) and the capacity of the four VLCCs then known to be available to MTI at Fujairah (1.4 million tons), the figures do not tally closely enough to persuade me that Mr. Heuzé was the source of Mr. Silverberg's information. On the other hand, a calculation based on the total capacity of the vessels at Fujairah available to MTI at that time would have produced a rather different figure. At the meeting in Athens on 7^th February Mr. Kilakos produced a list of vessels and cargoes held in by MTI on 6^th February which had been drawn up at his request by the master of the Metrotank, Capt. Kapsomenakis. This indicates that the total capacity available to MTI at the time was something in excess of 2 million tons. However, not all the vessels on Capt. Kapsomenakis' list could properly be regarded as storage vessels in anything more than the most temporary sense. Indeed, some of them were about to leave Fujairah. It is quite possible, therefore, that if Mr. Kilakos did provide some information to Mr. Silverberg about the capacity of the storage vessels, he did not include all the vessels which found their way into Capt. Kapsomenakis' list.
224. Mr. Hawkins' spreadsheets enabled Glencore to make an accurate calculation of the total quantity of oil delivered to MTI. I think there is little doubt that the figure of 1.1-1.2 million tons was obtained by a rough comparison between the total quantity of oil delivered into store and the capacity of the storage vessels at Fujairah as known to Mr. Silverberg on 5^th February. On balance I think it unlikely that the information he had about the capacity of the storage facility came from Mr. Heuzé, not only because Mr. Heuzé was abroad at the time, but also because the total capacity of the vessels mentioned in his notebook would have suggested that the shortfall was rather greater than 1.1-1.2 million tons. I think the more likely answer to this conundrum is that some information was indeed provided by MTI by telephone which indicated that there was rather more storage capacity available than had been known to Mr. Heuzé.

The meeting of 7^th February

225. The meeting which took place in MTI's offices in Athens on 7^th February was attended by Mr. Silverberg, Mr. Heuzé and Mr. Palacios on behalf of Glencore and Mr. Kilakos and his son on behalf of MTI. Mrs. Gene was also in the offices during the meeting, but remained in another room except when she was asked to bring papers into the meeting and did not attend the discussions. The meeting began in the morning and continued throughout much of the day.
226. The atmosphere during the morning was described by all those present as "cool" and "professional". Although both Mr. Kilakos and his son were under some strain, the parties were able to address the situation in a relatively dispassionate way and set about establishing how much oil remained at Fujairah, how much had gone missing and the extent of MTI's liabilities, including its liabilities to the banks. With the help of his son Mr. Kilakos produced a number of typed sheets describing different aspects of the current position. These included a table showing the quantities of oil held on different vessels at Fujairah. Mr. Kilakos and Mr. Silverberg then set about identifying how much oil remained in storage and how much according to his view of the matter was missing.
227. On one of the sheets headed "Cargoes loaded (b/l issued) not yet invoiced" Mr. Silverberg made various notes. Under the heading "Missing" he listed quantities of fuel oil, gas oil and naphtha, drawing a distinction between oil which had been "ITT'd" (i.e. which had been made the subject of an ITT contract) and oil which was "out", i.e. for which no ITT contract had by then been issued. He recorded the total quantity missing as 1.374 million tons. A similar distinction was drawn by Mr. Palacios in a note that he made, also in the course of the meeting. He deducted the quantity of fuel oil remaining in store from the quantity not ITT'd and noted the difference as "Missing". He then noted separately the quantity which had been ITT'd but for which no letter of credit had been opened. Mr. Palacios, who was new to the joint venture arrangements, attempted to note the discussion between Mr. Silverberg and Mr. Kilakos, although he did not fully understand why it was being approached in that way.
228. Mr. Silverberg said that they had approached the calculations in the way I have described simply because it was easier to do so. Once they realised that oil which had not been sold to MTI under ITT contracts was missing, he said, it was safe to assume that all the oil which had been sold was also missing. I do not find that very convincing. The way in which a distinction was explicitly drawn between oil which had been the subject of ITT contract and oil which had not certainly suggests that Glencore regarded that as a significant distinction, as of course it would be if oil which had been the subject of an ITT contract was regarded as having been released to MTI on the ITT date. In those circumstances Glencore's interest in that oil, limited as it was to the retention of title, was essentially one of security for the payment of the price. Although in one sense "missing", it could not be regarded as having been wrongfully removed in the same way as oil which had not been sold to MTI at all. I think that is the most likely explanation for what one sees in these notes.
229. There was also an attempt during the meeting to assess MTI's outstanding financial obligations. A rough calculation indicated that they amounted in round terms to $194 million overall, as reflected in Mr. Silverberg's notes. That figure was made up of $177 million (an earlier figure of $175 million was overwritten) in respect of oil covered by existing ITT contracts or already taken from storage and a further $17 million owed to the banks.
230. Mr. Kilakos' account of this part of the meeting was somewhat confused. He said that the figure of $175 million (he did not comment on the overwritten figure of $177 million) represented the totality of MTI's obligations to the banks and all the oil claimants, including Glencore, not just the amount it owed to Glencore. It had been brought into the discussions, he said, because Glencore had indicated a willingness to support MTI and therefore needed to know the full extent of its liabilities. However, when pressed to explain how he had calculated the amount due to all these parties he failed to give an intelligible explanation. In my view the notes made by Mr. Silverberg during the meeting speak for themselves.
231. It was largely common ground that towards the end of the morning the meeting broke up to enable each side to consider its position. During that break Mr. Palacios made amendments to the draft agreement which he had brought with him on his computer. In the final version of the agreement the quantity of 1.374 million tons of oil which Mr. Silverberg had calculated to be missing was expressed as approximately 1.4 million tons. Mr. Kilakos said that he understood the figure of 1.4 million tons of missing oil to represent the amount of money which MTI owed Glencore, but have to say that I found his evidence on this point very difficult to understand and again, I think the documents speak for themselves.
232. The meeting resumed during the afternoon when the agreement in its amended form was produced and signed by Mr. Kilakos and his son. The course of the meeting was investigated quite carefully with all those who attended it because it was MTI's case that the signatures of Mr. Kilakos and his son had been procured by duress. There is no doubt that Mr. Kilakos was under great commercial pressure at this meeting because the future of the enterprise which he had built up and in which he had devoted so much effort, as well as his personal reputation, was at stake. He hoped that Glencore would stand by him and help him in a way which would enable MTI to survive (and so had an incentive to be co-operative rather than obstructive) but I do not think that he can have been under any illusions about the possible consequences if he was unable to reach an accommodation with Mr. Silverberg. Those who were present at the meeting gave rather differing accounts of what had gone on. Mr. Kilakos said that he had been threatened, but that was denied by Glencore's witnesses. In particular, Mr. Palacios, perhaps the most straightforward and reliable of all the witnesses in the case, was adamant that no threats had been uttered, or indeed that there had been any overt discussion of the consequences of a failure to sign. It is true that he had been out of the room on occasions making changes to the draft, but if threats of the kind suggested by Mr. Kilakos had been made, I think it unlikely that they would have escaped his attention entirely. I am quite satisfied that Mr. Kilakos signed the agreement voluntarily, although he found it unpalatable to do so. He was aware that MTI's financial position was very grave and he hoped that by co-operating with Glencore he would obtain a lifeline which would enable it to survive. John Kilakos signed it because his father asked him to do so and because he too was aware that it was the only hope for MTI. In the light of the evidence as a whole Mr. Crane very properly accepted that the allegation of duress could not be sustained.

The state of the inventory on 7^th February 1998

233. The agreement records that MTI "cannot account for" 1.4 million tons of oil delivered to MTI to be held to Glencore's order. The expression is a broad one and is capable of covering both an inability to redeliver oil from storage and an inability to pay for oil already taken from store. It was chosen when the draft was produced, as Mr. Palacios explained, in order to avoid more inflammatory expressions. Mr. Heuzé recalled that Mr. Kilakos had said on more than one occasion both prior to and during the meeting on 7^th February that MTI had done something very wrong in taking oil from the joint venture stock without Glencore's knowledge, though he also recalled that on one occasion Mr. Kilakos had said that he thought MTI was entitled to take oil first and pay for it later. Mr. Kilakos himself did not deny that he felt that MTI had let Glencore down badly by becoming unable to pay for the oil which it had used, though he continued to maintain that he believed MTI was entitled to draw on oil in storage even before it had become the subject of an ITT contract. Although I have been unable to accept his evidence in all respects, I found Mr. Heuzé generally a careful and reliable witness whereas there were many occasions on which I found Mr. Kilakos' evidence both confusing and evasive. In the light of the findings I have already made an admission on the part of Mr. Kilakos that MTI acted improperly in drawing oil from the joint venture stock in advance of an ITT contract does not carry the matter very far. Nonetheless, I think it right to record that I prefer the evidence of Mr. Heuzé on this issue. I do so partly for the reasons just given and partly because I am unable to accept that Mr. Kilakos really did believe that MTI was entitled to take oil before it had become the subject of an ITT contract. Nothing of that kind had been agreed between him and Mr. Heuzé, nor do MTI's own records reflect any such understanding. I am satisfied that in the highly-charged atmosphere which existed on 7^th February Mr. Kilakos did admit that he, and through him MTI, had acted wrongly.

Title to blended oil

234. Some of the fuel oil held in store by MTI on 7^th February may have been produced by blending parcels drawn from joint venture stocks in respect of which no ITT contracts or letters of credit had been issued. That is not something that can be finally determined at this stage, but it is necessary for me to deal with the question of principle which blending of that kind raises.
235. Under the law of Fujairah the blending of oils of different grades to produce a new product would be considered to involve the destruction of the original components and as a result property in the blended product would vest in the person creating it. However, if the parties had expressed an intention that property in the blended product was to vest in one or other of them, the law of Fujairah would respect and give effect to that intention: see paragraphs 89 and 108 of the judgment in Phase 1. Mr. Schaff submitted that in the present case it was possible to infer from the terms governing the parties' relationship under JV2 an intention on the part of both Glencore and MTI that property in any blended products produced by MTI using components drawn from the joint venture inventory for which no letter of credit had been opened should remain the property of Glencore. He also submitted that MTI could not rely on its own improper act as the basis for acquiring title to oil which would otherwise belong to Glencore. Mr. Crane, however, submitted, that Glencore had effectively acquiesced in the use of oil by MTI for producing blended product without making any agreement under which it would retain or acquire title in that product. Alternatively, if MTI had been acting wrongfully in using joint venture oil for blending, the parties had never given any thought to the possibility that MTI might deal with oil in storage in an improper way and therefore had not directed their minds to the consequences. In those circumstances it could not be said that they had formed any common intention about which of them should own the resultant blend so that the general rules of law applied.
236. The starting point for any discussion on the proprietary effects of blending is the view taken by the law of Fujairah that the blending operation involves the destruction of the original components and the creation of a new commodity. This naturally leads to the conclusion that property in the new commodity vests in the manufacturer, subject to any agreement on his part that it should vest in someone else (as would be the case, for example, under a contract for work) or to any rule of law to the contrary. However, the general law of Fujairah does not prevent the manufacturer of a new commodity from acquiring title to it simply on the grounds that in producing it he has wrongfully made use of property belonging to someone else. In order to displace the general rule, therefore, it is necessary in a case such as the present for anyone who claims that he has a better title to the goods to demonstrate the existence of an agreement under which title in the new commodity was to vest in him rather than in the manufacturer.
237. The fact that there is a pre-existing relationship between the owner of the goods and the wrongdoer who makes use of them for his own purposes is obviously an important part of the background when it comes to deciding whether there was any implicit agreement between them relating to matters of this kind. That may be particularly important in a case where, as here, one party deposits goods with another on terms that he will transfer the property in them to that other under certain conditions, but not otherwise. Mr. Schaff submitted, in effect, that an agreement in those terms amounts by implication to an agreement that property in the goods shall not pass to the bailee unless and until the relevant conditions are satisfied, and that that in turn amounts to an agreement that property in any new commodity manufactured in whole or part from those goods shall not vest in the bailee either.
238. This is a powerful argument, but I do not think that an agreement on the part of an owner of goods which have been deposited with a bailee to transfer the property in goods to the bailee subject to certain conditions can be elevated into an agreement that title in any new commodity manufactured from them by the bailee without the owner's consent shall vest in the original owner. The difficulty lies in the view taken by the law of Fujairah of the effect of the blending process. Once one accepts that it results in the destruction of the original constituents one is faced with the fact that something quite new has come into existence. All that an agreement restricting the conditions under which title in the original goods may pass to the bailee can achieve is to render the act of the bailee wrongful; it does not purport to deal with the vesting of title in the new commodity and I do not think that any such agreement can be inferred from it. The evidence of the experts on Fujairah law on this point at the Phase 1 trial proceeded largely on the assumption that the new commodity had been produced by a bailee to whom goods had been entrusted for safe keeping, as did the submissions based on it. There was no suggestion that the existence of an agreement to sell the goods to the bailee subject to certain conditions might affect the position and as a matter of principle I cannot see that it should do so.
239. Mr. Schaff sought to rely on the principle in Alghussein Establishment v Eton College [1988] 1 W.L.R. 587 in support of a submission that the contract between Glencore and MTI should not be construed in such a way as would enable MTI to benefit from its own wrongdoing. However, I do not think that Glencore can derive any assistance from that decision in the present case. It is important to remember that English law has very limited relevance in relation to questions of title to goods situated in Fujairah. In the present case it is confined to the construction of the contract insofar as that is necessary to ascertain the intention of the parties with regard to the vesting of title in the blended product. However, in Alghussein v Eton Lord Jauncey was of the view that in general the principle in question is embodied in a rule of construction rather than one of law and I therefore approach the question in the same way.
240. As applied by the House of Lords in Alghussein v Eton the rule takes the form of a presumption that the contract in question is not intended to be construed in a way which would allow one party to it to take advantage of his own breach. Usually that would involve one party's seeking to treat his own wrongful act as bringing into effect a provision of the contract in his favour. However, the present case is not of that kind at all. MTI does not seek to invoke any provision of the contract in its favour and it is difficult therefore to see how this principle of construction can be invoked to circumvent the language of the contract. But the matter goes further than that. The contract in the present case deals only with title to oil held in storage and MTI does not contend that title to that oil was transferred to it. Rather it maintains that title to that oil was destroyed and that it acquired an original title in a new commodity. This may appear to be a narrow distinction, but it is of critical importance in this case. In Alghussein v Eton Lord Jauncey left open the possibility that there may be a substantive rule of English law which prevents a person from benefiting from his own wrong, but for the reasons given in the Phase 1 judgment the substantive rules of English law have no relevance to this question.
241. It follows that in the case of parcels of fuel oil blended by MTI for its own purposes using oil drawn from the joint venture stock in respect of which no ITT contract had been issued property in the blended product vested in MTI under the general law of Fujairah. That is also the position in the case of parcels blended using oil in respect of which ITT contracts had been issued but no letter of credit had been opened. Oil covered by ITT contracts was by agreement between the parties treated as having been delivered to MTI with consent for its disposal in the ordinary course of business, including blending. Glencore's reservation of title could not survive the destruction of the oil brought about by the process of blending and for the reasons given above I do not think that one can find in the contract an implied agreement that title to blended product was to vest in Glencore. A fortiori title to blended products produced from oil in respect of which a letter of credit had been opened also vested in MTI. Only in the case where cargoes or parcels of fuel oil were blended from joint venture stock for Glencore to lift or retain in stock as part of the joint venture inventory did title to the blended product remain in Glencore.

The position of the Shipowner defendants

242. Among the claims made by Glencore in this litigation are claims against the owners of the Shoko and the owners of the Cherry, each of whom are alleged to have converted oil belonging Glencore. In addition to other issues raised in their defence, both owners put Glencore to proof of its title to the oil and also contends that whether or not title to the oil remained in Glencore at the relevant time, MTI had authority to deal with it and could therefore properly ship it on board their vessel. In due course it will be necessary to make specific findings of fact in relation to the shipments of oil on those two vessels, but for the purposes of this Phase I am concerned only with the question of Glencore's title to the oil in question and MTI's authority to deal with it, both of which arise in the context of Glencore's dispute with MTI. In his final speech, therefore, Mr. Goldstone on behalf of the shipowners adopted the submissions made by Mr. Crane. There was one additional matter, however, on which he made separate submissions, namely, the effect of any ambiguity in the scope of the authority given to MTI to deal with joint venture stocks.
243. Mr. Goldstone submitted that Mr. Kilakos genuinely believed that under the JV2 arrangements MTI was entitled to deal with joint venture stock from the time of its arrival at Fujairah and that if there was any ambiguity in the arrangements made between him and Mr. Heuzé at Singapore in June 1996 it should be resolved against Glencore. This submission was based on the authority of Ireland v Livingstone (1872) LR 5 HL 395 in which the House of Lords held that where a merchant's letter to a commission agent was worded so as to be fairly capable of two interpretations the merchant was bound to accept the purchase made by the agent who had reasonably and honestly acted on one of those meanings, even though it was not the one that the merchant had actually intended.
244. Mr. Goldstone's submission depends first of all on showing that the terms of the agreement between Mr. Heuzé and Mr. Kilakos at Singapore were fairly capable of two interpretations. In support of that he relied in particular on the admitted fact that it was agreed between them that MTI alone would be responsible for marketing the oil. He submitted that the agreement in those terms should be construed as giving MTI the right to sell and dispose of the oil brought into Fujairah pursuant to that agreement at its discretion. If it were not construed in that way it was redundant because Glencore, which would receive a fixed profit in any event under the new arrangement, had no interest in the marketing of the oil at all. Mr. Goldstone submitted, however, that whether that was ultimately right or wrong, it was a view of what had been agreed which could fairly and honestly be held, and was held by Mr. Kilakos. That being so, Glencore could not treat MTI's disposal of the oil pursuant to the agreement as unauthorised.
245. Ireland v Livingstone is a rather unusual case. It is apparent from the speech of Lord Chelmsford that the wording of the letter in question did not merely raise a doubt, but was genuinely ambiguous in that it was, in his view, equally capable of bearing two different meanings. In most cases, however, it is possible to identify the correct meaning of a document, even though the wording used may at first sight give rise to some doubt. The difficulties of applying this principle to the present case are compounded, however, by the fact that the court is not being asked to construe a document but to ascertain the meaning to be ascribed to a series of conversations, none of which either Mr. Heuzé or Mr. Kilakos could recall in reliable detail and none of which were recorded in any form of contemporaneous note. The problem, therefore, is not so much in the construction of a particular form of words but in ascertaining what it was that the parties agreed. For the reasons I have already given I am unable to accept that it was agreed between them that MTI should be entitled to dispose of oil before any ITT contract had been issued in respect of it. I do not consider that there was any ambiguity about the agreement in this respect or that Mr. Kilakos believed that the new arrangements involved a fundamental change in the parties' relationship of that kind. I therefore reject Mr. Goldstone's submission.

The Fal defendants

246. The claims being made by Glencore against Fal Shipping Co. Ltd and Fal Oil Co. Ltd ("the Fal defendants") relate to four shipments. Three of these took place before the end of January 1998 and raise the same issues as those which arise in relation to other cargoes. The fourth, however, took place on 8^th-9^th February 1998 after MTI had entered into the agreement with Glencore on 7^th February and therefore raises rather different issues.
247. Between 1930 hours on 8^th and 1736 hours on 9th February 1998 MTI shipped a cargo of 24,415 tons of low sulphur fuel oil on board the vessel Horizon XII under a contract of sale made with Fal Oil Co. Ltd in December 1997. A bill of lading providing for carriage to Sharjah was issued on MTI's standard form naming MTI as shipper and BTCL as consignee. Glencore's claim in respect of this cargo is pleaded in conversion, but for the reasons given in the judgment in Phase 1, issues relating to the wrongful interference with oil held in the floating storage are governed by the law of Fujairah. Under Article 1325 of the Civil Code sale and delivery to a bona fide purchaser would be sufficient to pass title, subject to the right of the owner to recover from the buyer under Article 1326 in circumstances where the goods had been misappropriated. The essential issue between Glencore and the Fal defendants, therefore, is whether the shipment of this cargo by MTI involved a misappropriation of Glencore's property or whether it was one to which Glencore had consented. Mr. Davey on behalf of the Fal defendants submitted that whatever may have been the position prior to 7^th February, everything changed once Mr. Kilakos signed the agreement placing the whole of MTI's operations under Glencore's control. He submitted that Glencore had been aware of the shipment on the Horizon XII at an early stage and, far from taking any steps to interrupt it, had allowed it to be completed and had subsequently taken control of the shipping documents in order to retain control over the cargo.
248. The agreement of 7^th February gave Glencore the right to monitor and approve the movement of oil into and out of the floating storage. It also gave Glencore complete control over MTI's commercial activities. One of the purposes of these provisions was to prevent any disposal of oil in which Glencore considered itself to have a proprietary interest without being able to ensure that it received the proceeds. The agreement contemplated that MTI would continue in business, but only under Glencore's supervision. As Mr. Silverberg confirmed, from Glencore's point of view it did not matter if oil held in storage were sold to third parties provided that the proceeds of sale remained under its control. It was for that reason that Glencore was willing to allow MTI to fulfil existing commitments under its direction.
249. On 8^th February there was a meeting at Glencore's offices in London to take stock of the position. Following that meeting Glencore sent messages to the masters of the storage vessels informing them that the oil on board had been transferred to it and seeking confirmation that they would follow Glencore's instructions in relation to its disposal. One such message was sent to the master of the Metrotank who replied early the next day that ship-to-ship transfers were currently going on with the Horizon XII and the Athenian Horizon. It is likely that this information was distributed to a number of people in Glencore's office, but no steps were taken at any stage to prevent the shipment from being completed.
250. On 9^th February Mr. John Garrett arrived in Fujairah. He had been sent out by Glencore to monitor shipments of oil and to arrange for the amount of oil held in the floating storage facility to be measured. He was asked by his superior, Mr. Bloss, to take charge of the shipping documents for the cargo on the Horizon XII and he took delivery of them from MTI on 10^th February. He delivered them to another employee of Glencore, Mr. Jan de Laat, for carriage to London. Beyond that, however, he played no part in the loading of the vessel.
251. At about the same time as Mr. Garrett was sent out to Fujairah another of Glencore's employees, Mrs Freeman, was sent to Athens to monitor operations in MTI's office there. She arrived there during the morning of 9^th February. On a copy of a telex from BTCL dated 6^th February advising MTI of the opening of a letter of credit covering 25,000 tons +/- 5% of low sulphur fuel oil she noted "Horizon 12 loading now", from which it seems clear that someone in MTI's office had informed her that the shipment was taking place.
252. The arrangements for this shipment must have been made well before the meeting of 7^th February, but I am unable to accept that this is a case in which the cargo was delivered to the defendants without Glencore's approval. Although at the time of the meeting on 7^th February Glencore had little idea of the extent of MTI's commitments, all those involved must have been aware, as indeed Mr. Heuzé recognised, that its current operations would continue unless steps were taken to interrupt them. One way of doing that would have been to send immediate instructions to the loading master at Fujairah and to the masters of the storage vessels to cease all operations pending further instructions from Glencore. Steps of that kind could have been taken, but for understandable reasons Glencore preferred to allow MTI's operations to continue while monitoring and controlling what went on. It was advised of the shipment on the Horizon XII and in due course took control of the shipping documents which would ordinarily have given it control of the cargo. The fact that even now no payment has been made for the cargo is not a matter which can have any bearing on the question I have to answer.

The Questions

253. The conclusions I have expressed in the course of this judgment and the reasons given for them do, I believe, provide answers to all the questions that were formulated for the purposes of this Phase of the litigation. However, the questions themselves are in some cases complex, and even where they are not, do not necessarily admit of a simple answer. I propose, therefore, to invite further submissions from counsel before deciding whether it would be of assistance to formulate specific answers to them, and if so, in what terms.
254. Finally, I do not think it would be right to leave this Phase of the Metro litigation without recording my admiration for the manner in which it has been prepared and conducted by all those involved and my gratitude for the assistance I have received.
Appendix

The questions which the court directed should be determined in Phase 2 of the litigation are set out in this appendix.

In the light of the parties' submissions following the handing down of judgment the questions should be answered in the manner indicated. They should be read and understood in the context of the judgment as a whole.

Question 1: Were the cargoes identified in schedule 1 to Glencore's Points of Claim the subject of the transactions set out therein.

Answer: Yes.

Question 2: Pursuant to what agreement(s) between Glencore and MTI and on what terms were the cargoes identified in schedule 1 to Glencore's Points of Claim delivered into and subsequently held in floating storage off Fujairah or otherwise dealt with?

Answer: (a) The cargoes were delivered by Glencore to MTI under the terms of an umbrella agreement, described by the parties as a 'joint venture agreement', made orally June 1996.

(b) The following, among others, were terms of the joint venture agreement:

(i) that Glencore would purchase cargoes of oil in accordance with the directions of MTI for delivery to MTI to hold in storage at Fujairah pending sale to MTI;

(ii) that MTI would buy from Glencore all the oil delivered into storage under the agreement;

(iii) that Glencore would sell oil held in store to MTI in parcels by in-tank transfer as requested by MTI from time to time.

(c) Glencore sold parcels of oil to MTI pursuant to the joint venture agreement under individual contracts of sale by in-tank transfer on the following, among other, terms:

(i) that property in the oil would pass to MTI on the issue by Glencore of a stock transfer certificate;

(ii) that MTI would be be free to dispose of the oil as from the time of delivery;

(iii) that delivery would take place on a date agreed between the parties, that being the date on which stock transfer would be deemed to have taken place for payment purposes.

Question 3: In relation to the cargoes delivered into and subsequently held in floating storage off Fujairah pursuant to the arrangements in paragraph 2 hereof,

(1) in relation to cargoes initially purchased by MTI from third parties as identified in Schedule 1 to the Points of Claim, did Glencore acquire title to such cargoes under valid and enforceable back to back sale and purchase contracts between itself as buyer and MTI as seller?

Answer: Yes

(2) in relation to cargoes initially purchased by Glencore from third parties as identified in Schedule 1 to the Points of Claim, did Glencore acquire title to such cargoes under valid and enforceable contracts between itself as buyer and such third parties as sellers?

Answer: Yes

Question 4: All issues as to the true construction of the relevant agreement(s) (and the express and/or implied terms thereof) between Glencore and MTI and in particular

(1) Blending

(a) whether the relevant agreement(s) expressly or impliedly provided that MTI was permitted to blend some or all of such cargoes, either with other oil belonging to Glencore or with other oil belonging to MTI or with other oil belonging to other persons, and if so with which other oil(s) and upon what terms;

Answer: The contracts made no provision for blending. However, as between itself and Glencore, MTI was permitted to blend oil delivered to it under contracts for sale by in-tank transfer with other oil belonging to MTI or with oil belonging to other persons.

(b) upon what terms as to title in the blended product(s) was MTI permitted to blend as set out above and, in particular, did the relevant agreement(s) expressly or impliedly provide that Glencore would retain or acquire title in the blended product or in a proportionate share of the blended product or that MTI would acquire and retain title in the entire share of the blended product;

Answer: It was implicit in the delivery of oil to MTI under an ITT contract with permission to dispose of it that title in any blended product should vest in MTI to the exclusion of Glencore.

(2) Commingling

(a) whether the relevant agreement(s) expressly or impliedly provided that MTI was entitled to commingle such cargoes with other oil belonging to Glencore or with other oil belonging to MTI or with other oil belonging to other persons, and if so, in what circumstances, in relation to what degree of similarity in terms of grade or specification and upon what terms;

Answer: Yes, provided only that the oil was segregated by grade in accordance with the definitions agreed between Glencore and MTI;

and if so,

(b) in what bulk(s), identified by reference to what identifying characteristics (such as grade, specification, location of storage or otherwise) did the relevant agreement(s) expressly or impliedly permit such commingled storage;

Answer: in relation to each grade the relevant bulk was the total quantity of that grade held in store at Fujairah by MTI;

(c) upon what terms as to title in any commingled bulk and, in particular, whether the relevant agreement(s) expressly or impliedly provided that Glencore would retain or acquire title in the commingled bulk or in a proportionate share of the commingled bulk or that MTI would acquire and retain title in the entire share of the commingled bulk(s);

Answer: It was implicit in the terms of the joint venture agreement that upon commingling in storage of oil owned by Glencore with oil owned by MTI or other persons Glencore would become an owner of the whole of the commingled bulk in common with MTI and any other persons whose oil had contributed to the bulk in proportion to the quantity contributed by each of them.

(3) Oil kept separately

Insofar as the oil the subject of such cargoes was kept separately by MTI (neither commingled nor blended), upon the true construction of the relevant agreement(s) did Glencore retain title in that oil, or did MTI acquire title in that oil and, if so, when and how?

Answer: MTI acquired title in that oil when, having been delivered to MTI under a contract of sale, it was disposed of by MTI in the ordinary course of business by delivery to a third party or by consumption in the production of a blended product.

Question 5: Upon the true construction of the relevant agreements as determined in 1 above, and in the light of the judgment on the phase 1 issues, in relation to those cargoes the subject of paragraphs 3(1) and 3(2) above in respect of which Glencore had acquired title and

(a) which were never the subject of any ITT contracts between Glencore as seller and MTI as buyer; or

(b) which were the subject of ITT contracts between Glencore as seller and MTI as buyer but which were never the subject of letters of credit and stock transfer certificates thereunder;

did property therein pass to MTI and if so when and how and to what extent?

Answer: (a) Property did not pass to MTI in any cargoes or parcels of oil which had not become the subject of ITT contracts between Glencore as seller and MTI as buyer;

(b) Property did pass to MTI in parcels of oil which were the subject of ITT contracts between Glencore as seller and MTI as buyer, despite the absence of letters of credit and stock transfer certificates, when, following delivery under the contract, MTI, disposed of the oil, whether by way of sale to third parties or in the production of blended products;

(c) Property in blended products produced by MTI using oil which had not become the subject of an ITT contract between Glencore as seller and MTI as buyer vested in MTI under the law of Fujairah.

Question 6: In relation to such cargoes the subject of paragraphs 3(1) and/or 3(2) above in respect of which Glencore had acquired and retained title, did MTI nonetheless have the actual (express or implied) authority of Glencore to sell and/or deliver such cargoes to third parties and/or did any such sale and/or delivery take place without Glencore's knowledge?

Answer: (a) MTI did not have authority to sell or deliver to third parties cargoes in which Glencore retained title, except in the circumstances identified in the answer to Question 2.

(b) Glencore was not aware of the sale and delivery by MTI of oil which had not been delivered to MTI under contracts of sale.

Question 7: In relation to the cargoes identified in Schedule 1 which were not delivered into and subsequently held in floating storage off Fujairah at all but were either retained on board the relevant carrying vessel or transhipped onto another carrying vessel for onward delivery to a third party, the best particulars of which are set out in Schedule 3 to the Points of Claim, to what extent (if at all) are the questions set out in paragraphs 2 - 6 above to be answered differently?

Answer: The questions are not to be answered differently.