England and Wales High Court (Commercial Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		04/02/2022

DAVID EDWARDS QC SITTING AS A DEPUTY JUDGE OF THE HIGH COURT

"Covid-19 Protocol: This judgment was handed down by the judge remotely by circulation to the parties' representatives by email and release to Bailii. The date and time for hand-down is deemed to be 04 February 2022 at 14:00"

David Edwards QC :

A. Parties

1. The Claimants, Provimi France S.A.S., Cargill S.L.U., Cargill Poland Sp. Zo.o and Provimi Limited were originally part of the Provimi group of companies. Save where it is necessary to distinguish between them, I will refer to the Claimants (and to the group generally) as "Provimi". Provimi was acquired by Cargill, Inc. in November 2011, and the Claimants are now part of the Cargill group which provides food, agricultural, financial and industrial products globally.
2. The Defendant, Stour Bay Company Limited ("Stour Bay"), is an English company that specializes in the supply and distribution of vitamins, minerals and amino acids for the animal feed, food and beverage industries. Stour Bay is a distributor: it does not manufacture these products itself but sources them from third-party manufacturers and intermediaries for on-sale to its customers.

B. Background

3. In February 2013, following discussions between the parties that first commenced in 2010, Provimi began purchasing a vitamin D3 product from Stour Bay. The particular purchases with which the present action is concerned were made in and after January 2015 under contracts of sale that I will describe later, but there were earlier purchases in 2013 and 2014.
4. The vitamin D3 product was sourced by Stour Bay from an Indian company, Fermenta Biotech Limited ("FBL"). The product was described in FBL's Product Information Sheet as "Vitamin D3 500 Feed Grade" and I will refer to it in that way in this judgment. Stour Bay accepts that it knew that the product was going to be used by Provimi to manufacture some form of animal nutrition product, although it says that it did not know precisely what product Provimi was going to make.
5. In the event, the Vitamin D3 500 Feed Grade product purchased by Provimi from Stour Bay was mixed by Provimi with trace elements and other ingredients to manufacture a poultry pre-mixture, which was then sold by Provimi to its customers - either feed mills or poultry farms – who then combined the Provimi pre-mixture with macro-nutrient ingredients to produce a poultry feed that was ultimately fed to chickens.
6. The purpose of adding vitamin D3 to poultry feed in this way is to support normal bone mineralization. The amount of naturally present vitamin D in animal feed ingredients is limited, and intensively reared chickens have limited access to sunlight and thus to a natural source of the vitamin. Vitamin D3 deficiency in chickens can result in rickets or osteoporosis in young growing birds and/or in poor eggshell quality.
7. Around the end of July 2015 (and possibly before then) Provimi began to receive complaints from some of its customers who reported low vitamin D3 levels in Provimi's pre-mixture and problems with their poultry, including poor growth, lameness, rachitis and abnormal mortality rates. The issue was investigated, and Provimi settled a number of claims for compensation.
8. It is common ground between Provimi and Stour Bay in these proceedings that the cause of the poor performance of the affected poultry was vitamin D3 deficiency. A number of other matters are also not in dispute, namely that:

i) The source of the vitamin D3 in the Provimi pre-mixture sold to the relevant customers was the Vitamin D3 500 Feed Grade product purchased by Provimi from Stour Bay and by Stour Bay from FBL;

ii) As at the date when the Vitamin D3 500 Feed Grade product was delivered by Stour Bay to Provimi, at that stage as a stand-alone product, the product contained fully compliant levels of vitamin D3; and

iii) The level of vitamin D3 found in certain batches of Provimi's pre-mixture produced in the spring and summer of 2015 had, however, dropped to unacceptable levels.

9. Thus, although the vitamin D3 content of the Vitamin D3 500 Feed Grade product was acceptable at the time the product was delivered to Provimi, thereafter, either as a result of being mixed with other ingredients during the course of producing the pre-mixture or for some other reason, the vitamin D3 content diminished or degraded, leading to vitamin D3 deficiency in the affected poultry.

C. The Proceedings

10. On 10 October 2019 Provimi commenced proceedings against Stour Bay alleging that the Vitamin D3 500 Feed Grade product that Stour Bay had sold to it was defective and claiming damages for breach of the contracts of sale.

(i) Provimi's claim

11. Provimi's pleaded case is that the content of the Vitamin D3 500 Feed Grade product sold to it and/or the way in which the product had been manufactured, in particular the absence of a gelatin or other coating, meant that the product was insufficiently robust and/or stable such that it could not withstand the normal processes involved in producing a poultry pre-mixture, including combination with other usual pre-mixture ingredients.
12. Specifically, Provimi alleges that:

i) It was an express term of the contracts of sale that the Vitamin D3 500 Feed Grade product would meet the requirements of a particular Provimi Ingredient Specification, identified by PAC Code 191910101, ("the Provimi Gelatin Specification") in terms, inter alia, of ingredients, composition, stability and content, in particular a requirement for a non-ruminant gelatin coating;

ii) There were, in addition, under sections 14(2) and (3) of the Sale of Goods Act 1979, as amended, ("the SGA"), implied terms that the Vitamin D3 500 Feed Grade product would be of satisfactory quality and reasonably fit for the purpose of adding to Provimi's and/or to an industry-standard pre-mixture, fitness for purpose in that context, Provimi says, requiring:

"… the Product to be sufficiently robust and stable that the poultry to which [the pre-mixture] would ultimately be fed would receive the envisaged and/or a sufficient level of Vitamin D3";

iii) The Vitamin D3 500 Feed Grade product did not comply with the Provimi Gelatin Specification and/or was not of satisfactory quality or reasonably fit for purpose in a number of respects:

a) It did not contain a coating of non-ruminant gelatin (or any coating) and it was not spray-dried;

b) It was manufactured using an atypical process, akin to a pharmaceutical granulation process, the result of which was that the vitamin D3 ingredient was unduly exposed to the aggressive processing conditions present in pre-mixture manufacture;

c) It did not contain 500,000 I.U./g of vitamin D3 upon or shortly after being combined with other ingredients in the pre-mixture; or, if it did it, it had a shelf-life which was not such as to allow the product to maintain such a vitamin D3 content; and/or

d) The product was, as a result of these matters, unstable and/or insufficiently durable, and the vitamin D3 in the product degraded upon or shortly after being mixed with other, ordinary ingredients that were included in Provimi's pre-mixture.

13. Provimi claims damages representing the amounts it says it paid to settle customer claims in France, Spain and Poland, some EUR 2,029,090.12. It also seeks to recover EUR 69,413.03 representing a refund it alleges Stour Bay agreed to make for Vitamin D3 500 Feed Grade product that was returned after the problem with the product materialised.

(ii) Stour Bay's defence

14. There are a number of strands to Stour Bay's defence.
15. First, Stour Bay denies that the Provimi Gelatin Specification formed part of the contractual terms. Stour Bay alleges, in that context, that:

i) Provimi had been told by Stour Bay in April 2010, and again in documentation provided in 2011 and 2012 as part of the material evaluation and supplier audit processes that I will describe later, that the Vitamin D3 500 Feed Grade product produced by FBL and sold to Provimi did not contain gelatin; and

ii) The absence of a gelatin or any other coating was or ought to have been apparent to Provimi in any event upon examination and when the product was tested and approved by Provimi's Quality Assurance department, which was a necessary step before any order for the product could be placed.

16. Secondly, Stour Bay asserts that, as a result of a course of dealing between the parties, the contracts of sale concluded in and after January 2015 incorporated Stour Bay's standard terms and conditions ("the Stour Bay T&Cs") which, it says, were routinely included on the reverse of the invoices that were sent to Provimi.
17. The incorporation of the Stour Bay T&Cs is said by Stour Bay to have three particular consequences:

i) Under clauses 7.1 and 7.6 Provimi relied upon its own judgment in deciding whether or not to use the Vitamin D3 500 Feed Grade product (including whether to mix it with other ingredients), and was obliged to indemnify Stour Bay in respect of any losses, liabilities or claims arising from or in connection with its use of the product;

ii) By reason of clause 10.3 (and as permitted by section 55 of the SGA) the implied terms of satisfactory quality and fitness for purpose under the SGA were excluded. Mr Kulkarni, QC, who appeared with Ms Honey for Provimi at the trial, accepted that, if (which he denied) the Stour Bay T&Cs were incorporated, this consequence would follow; and

iii) In accordance with clauses 10.1,10.5.1 and 10.5.2 Stour Bay's only obligation to Provimi was to endeavour to assign to Provimi any rights that it had against FBL; Stour Bay's liability to Provimi, if any, was, in any event, limited to the contract price. Again, it was accepted by Mr Kulkarni, QC that, if the Stour Bay T&Cs were incorporated, these limitations would apply.

18. Thirdly, Stour Bay contends that, even if the Stour Bay T&Cs were not incorporated into the contracts of sale and the SGA implied terms were, therefore, not excluded, Stour Bay was nonetheless not in breach of those terms. Ms Ansell, QC, who appeared with Mr Thorne for Stour Bay, relied in particular upon section 14(2C) of the SGA and the exception concerning reliance set out in in section 14(3).
19. Fourthly, Stour Bay denies that any defect in the Vitamin D3 500 Feed Grade product was the cause of the problems that occurred in the poultry in 2015. In that context:

i) Stour Bay points to the fact that the Vitamin D3 500 Feed Grade product had been supplied to and used by Provimi apparently (though this was not accepted by Provimi) without incident or complaint during 2013 and 2014; and

ii) Stour Bay says that the true, and the sole effective, cause of the problems experienced in 2015 was the heatwave that occurred in continental Europe during the summer of that year, it being common ground between the experts that excessive heat is capable of affecting the susceptibility of vitamin D3 to degradation.

20. Fifthly, so far as the quantum of Provimi's claim is concerned, in paragraph 22 of its Defence Stour Bay denied Cargill's claim but pleaded no positive case of its own. In her written opening and closing submissions, however, Ms Ansell, QC, advanced two points:

i) She said that Provimi had failed to notify its customers of the problems with the pre-mixture, and had failed to withdraw the pre-mixture from the market, as soon as the vitamin D3 degradation had been discovered, and that Provimi had, therefore, caused and/or failed to mitigate its own losses; and

ii) She asserted that the burden lay on Provimi to prove the reasonableness of its settlements and that it could not do so; in particular, Ms Ansell, QC said, because some of the Claimant companies had failed to rely upon their own standard terms and conditions to remove or reduce their liability.

Provimi raised a pleading point in relation to these two matters, which I will address later.

21. Sixthly and finally, in relation to the claim to recover the purchase price for returned product, Stour Bay says that, not only has the supposed agreement between the parties not been pleaded properly, but there is also no adequate evidence of any such agreement. This aspect of the claim was, indeed, only lightly pressed by Mr Kulkarni, QC in his closing submissions.

D. The Issues

22. As will be apparent from the paragraphs above, the issues before me fall broadly into three categories:

i) Terms and Conditions: these issues include the question of whether the contracts of sale required compliance with the Provimi Gelatin Specification and whether the Stour Bay T&Cs were incorporated; but, as emerged in the course of submissions, there is, or at least there may be, an anterior issue as to what the relevant contract or contracts were and how they were made;

ii) Quality: this category embraces the issues concerning the quality of the Vitamin D3 500 Feed Grade product sold by Stour Bay, whether it breached the Provimi Gelatin Specification (if that formed part of the contracts of sale) and the SGA implied terms (if they were not excluded), and also the cause of the vitamin D3 degradation and the adverse effects in the poultry in 2015; and

iii) Loss and damage: whether Provimi can recover the settlements with its customers and/or whether it is entitled to recover the purchase price for returned product.

23. I will address the issues under these three headings when I come to address the parties' respective cases later in this judgment.

E. The Evidence

24. In addition to the documentary evidence included in the trial bundles that I was shown, I heard from 11 factual witnesses on behalf of Provimi and four factual witnesses on behalf of Stour Bay. The Provimi witnesses all lived and worked abroad; the Stour Bay witnesses did not. As a result of the Covid-19 pandemic, however, all the witnesses of fact from both parties gave their evidence remotely.
25. On the Provimi side, the contracts with Stour Bay were negotiated by Provimi's central purchasing department in the Netherlands (part of Provimi B.V.) ("Provimi Netherlands"), and I heard from Pierre Piccolin and Loic Amouroux, the two individuals involved. Martin van der Eijk, who worked in Quality Assurance at Provimi Netherlands gave evidence about a material evaluation of the Vitamin D3 Feed Grade product and a supplier audit of FBL which were carried out in 2011 - 2012.
26. Other Provimi employees gave evidence about the purchases and the use of the Vitamin D3 500 Feed Grade product by particular Provimi entities in particular countries, the complaints made by customers, the investigations that were carried out to try to identify the cause of the problem, and the settlements of customer claims. Specifically:

i) Benoit Chambon was employed by the First Claimant ("Provimi France"). He gave evidence about a number of general matters, but specifically about quality control and the pre-mixture manufacturing process at Provimi France. Erik Benard, an in-house lawyer at Provimi France, addressed the settlement of three claims made against Provimi France;

ii) Maria Sanz, Diego Valencia and Jorge Martinez gave evidence about the purchasing and the pre-mixture manufacturing process at the Second Claimant ("Provimi Spain") and about complaints received by Provimi Spain during 2015, including one "official" complaint from Huevos-Leon to whom Provimi Spain paid compensation; and

iii) Pawel Fiedorow was employed by the Third Claimant ("Provimi Poland") but also had a global role. He dealt with the use of the Vitamin D3 500 Feed Grade product in Poland and with the investigation of customer complaints. Pawal Kala dealt with the ordering process. Rafal Wysmolinski addressed the settlement of a claim by Sylwia Chibowska-Potega ("Syl-Drob").

27. So far as Stour Bay is concerned, I heard evidence from Nicholas Gibbons, its Managing Director and, along with Lorraine Jackson, one of the two founders of the company, who dealt with the negotiation of the contracts with Mr Piccolin and with the purchase of the Vitamin D3 500 Feed Grade product from FBL. Ms Jackson also gave evidence in relation to some of these matters.
28. I also heard evidence from Gillian Pickford and Alison Morgan, who were involved in the administrative side of Stour Bay's business, in Ms Pickford's case in making shipping arrangements and in Ms Morgan's case in preparing and posting invoices, each invoice, according to Stour Bay, either having the Stour Bay T&Cs printed on the reverse or being accompanied by a copy of them.
29. I will refer to the evidence of the factual witnesses as appropriate in the course of this judgment. In general, I was satisfied that all the witness gave their evidence honestly and with the aim of assisting the court, although in one or two cases their oral evidence appeared from time to time somewhat coloured by the legal arguments their respective companies were advancing, a point that I have taken into account.
30. As in many commercial cases, particularly a case like this that involves events taking place between six and twelve years ago (between 2009 and 2015), I consider that the surest guide to the facts is the contemporaneous documentation, the inferences that can properly be drawn from it, and the inherent probabilities. There were, however, areas where relevant documents were absent.
31. So far as that is concerned, Stour Bay complained about a substantial lacuna in Provimi's disclosure. The relevant facts concerning the complaint are as follows:

i) Disclosure was given by the parties by reference to the Disclosure Pilot in CPR PD 51U. Paragraph 3 of the Pilot obliges a person who knows that it is or may become party to proceedings to take reasonable steps to preserve documents within its control, imposing an obligation on legal representatives to take reasonable steps to advise a party to comply with its Disclosure Duties;

ii) The Claimants were on notice of the potential for litigation by no later than the end of 2015 – as explained below, a letter of claim was sent by Provimi to Stour Bay in early January 2016. The Claimants' current solicitors, Pinsent Masons LLP ("Pinsent Masons"), were instructed by no later than 28 April 2016 when they wrote to Stour Bay's solicitors, Clyde & Co LLP ("Clydes");

iii) On 2 December 2020, shortly before the disclosure deadline, Pinsent Masons advised Clydes that, in accordance with a Cargill document retention policy, documents held in individual Microsoft Outlook files had been permanently deleted after three years, and accordingly that Outlook documents pre-dating 2016 – the entirety of the period relevant to the claim – may not be available.

32. Subsequent correspondence between the solicitors suggests that the failure to put in place a litigation hold to prevent documents from being automatically deleted was not the result of Pinsent Masons failing to give advice, but the result of a Cargill in-house lawyer, apparently unfamiliar with procedure in common law jurisdictions, misunderstanding or failing to appreciate the advice given as to the types of documents that should be retained.
33. However it arose, the result - the deletion of emails over an entire period which are likely to have included relevant documents - is highly regrettable. Insofar as there was a misunderstanding, it serves to emphasise the importance of solicitors dealing with clients unfamiliar with English disclosure rules explaining the position fully. It may well be that an instruction simply to retain relevant documents, without explaining or ensuring that the client understands exactly what "relevant" means, is not enough.
34. I have borne in mind, when considering the evidence, that Stour Bay may have been deprived of access to documents potentially supportive of its case and/or potentially adverse to Provimi's case that would have been available to it if an appropriate litigation hold had been put in place as it plainly should have been. In considering whether I should draw an adverse inference, either from the absence of documents or, as Stour Bay also suggested, from the absence of relevant witnesses, I have followed the approach of Lord Leggatt JSC in Efobi v Royal Mail Group Ltd [2021] UKSC 33, [2021] 1 WLR 3863 at [41] where he suggested that the answer is largely a matter of ordinary rationality and common sense.
35. In addition to the factual evidence, I also heard from two expert witnesses, David Pickard on behalf of Provimi and Professor Colin Whitehead on behalf of Stour Bay. Both experts gave evidence in relation to animal feed additives and pre-mixtures and the likely cause(s) of the problems experienced during 2015. Both were well-qualified to give the evidence they gave, although their backgrounds were somewhat different:

i) Mr Pickard trained as a research biologist, but he had worked in the animal feed industry for most of his career in technical and regulatory roles, including a spell working for Lohmann Animal Nutrition GmbH, a manufacturer and supplier of vitamins for animal feed use, including stabilized formulations of vitamin D3;

ii) Professor Whitehead, in contrast, had a largely academic background. Prior to his retirement, he was a Professor at the Roslin Institute in Edinburgh, one of the world's leading institutes for animal science research. He had also served on a number of national and international bodies concerning poultry science.

36. I was assisted by the experts, who both gave their evidence fairly and straightforwardly, consistent with their duties to the court. Both experts were impressive in their own way, and both were realistic as to the conclusions that could or could not properly be drawn from what was known and the extent to which the questions they were asked were really matters of speculation. As I will explain, there was, in fact, a good deal of common ground between them.

F. The Facts

(i) The relationship between Stour Bay and Provimi

37. The relationship between Stour Bay and Provimi (within which I include Provimi's predecessor companies SCA Nutrition, SCA Nutec, Trouw Nutrition Nutec and Ireland Mixrite) commenced in around 2009.
38. The documents suggest that the first contract of sale between Stour Bay and Provimi was made in October 2009 when Mr Gibbons agreed with Mr Piccolin the sale of a quantity of choline chloride, a feed additive, to SCA Nutec (which became Provimi UK). Mr Gibbons also referred in his witness statement to the sale of a quantity of ammonium chloride to SCA Nutrition (which became Provimi Ireland) around the same time.

(ii) Invoices and the Stour Bay T&Cs

39. The first invoice rendered by Stour Bay to Provimi appears to have been invoice 3912 which related to the 2009 choline chloride shipment referred to in paragraph 38 above. Although it was mentioned in email exchanges between the parties in February 2010 when payment was being chased, a copy of this particular invoice did not appear to be in the trial bundles.
40. The bundles did, however, include a copy of invoice 4042 dated 3 March 2010, which was an invoice for a shipment of choline chloride to SCA Nutrition. The invoice was printed on Stour Bay headed notepaper, which bears a colourful animal logo on its front. There was no reference to them on the front, but printed on the reverse of the invoice were the Stour Bay T&Cs. The Stour Bay T&Cs also appeared on the reverse of invoices 4302, 4245 and 4281, three further invoices that were issued to Provimi in respect of choline chloride shipments in July, October and November 2010.
41. The Stour Bay T&Cs, which referred to Stour Bay as the "Seller" and to the third-party purchaser as the "Buyer" include the following provisions:

"2. CONDITIONS

2.1 The Contract shall be on these Conditions to the exclusion of all others (including any terms or conditions which the Buyer purports to apply).

2.2 Any quotation is issued by the Seller on the basis that no Contract will come into existence until the Seller despatches an acknowledgment of order (an 'Order Acknowledgment') to the Buyer.

2.3 So far as they are not expressly varied in writing (any such variation shall be ineffective unless it be approved by a director of the Seller), these Conditions shall apply to all Contracts and all Goods supplied and Services undertaken by the Seller and a Contract may only be cancelled or varied with the Seller's written consent on terms that the Buyer will indemnify the Seller against all losses incurred by it as a direct consequence of such cancellation.

…

4. DELIVERY

…

4.6 Where the Buyer may call for delivery of the Goods ('call off') by or at particular dates or over a stated period then (a) notwithstanding the terms of any call off by the Buyer, the Buyer is obliged to purchase all the Goods; (b) time shall be of the essence in respect of the Buyer's obligation to call off each instalment of the Goods at the date(s) or within the period(s) stated in the Order Acknowledgment; (c) deferral of any particular call off agreed by the Seller shall not relieve the Buyer of its obligation to make timely call offs of other instalments of the Goods; and (d) without limitation to Seller's other rights, if Buyer fails to call off any of the Goods in accordance with the terms stated in the Order Acknowledgment the Seller may in its discretion require the Buyer to call off and pay for any or all of the Goods forthwith or at such time or times as the Seller decides.

…

7. RISK OF USE; HEALTH AND SAFETY

7.1 As used in these Conditions 'use' of the Goods includes (without limitation) consumption, conversion, testing, mixing with other goods, fabrication, processing or other application, alteration, packing, storing or movement; 'disposal' includes (without limitation) sale, hire, consignment, pledge, gift, delivery, release or other disposal or encumbrance of the goods or any product or waste derived from the Goods whereby the Goods cease to be owned and in the exclusive possession of the Buyer.

…

7.6 Buyer acknowledges that in deciding whether or not to buy, use or dispose of the Goods or any part of them the Buyer relies on its own judgment and evaluation and its own examination and testing of the Goods and not on any representation warranty or advice or recommendation from Seller. Buyer agrees to indemnify Seller against all losses, liabilities or other claims (including without limitation third party claims) arising from or in connection with any use or disposal of the Goods by Buyer or by any person obtaining the Goods from Buyer (including but not limited to any claims attributable to defective goods or services).

…

10. LIMITATION OF LIABILITY

10.1 In respect of Goods not manufactured by the Seller, the Seller makes no representation and gives no warranty in respect of the sources or origin of manufacture or production of the Goods or any part thereof, but the Seller shall endeavour to assign for the benefit of the Buyer such rights (including guarantee or warranty rights) as the Seller has against such manufacturer but shall not be liable for such Goods beyond this Condition 10.1.

10.2 Subject to Condition 10.1, the following provisions set out the entire financial liability of the Seller (including any liability of the acts or omissions of its employees, agents and sub-contractors) to the Buyer in respect of any breach of these Conditions and any representation, statement or tortious act or omission, including negligence arising under or in connection with the Contract.

10.3 All warranties, conditions and other terms implied by statute or common law (save for conditions implied by section 12 of the Sale of Goods Act 1979) are, to the fullest extent permitted by law, excluded from the Contract.

…

10.5 Subject to Conditions 10.3 and 10.4:-

10.5.1 the Seller's total liability in contract, tort (including negligence or breach of statutory duty), misrepresentation or otherwise, arising in connection with the performance or contemplated performance of the Contract shall be limited to the Contract price; and

10.5.2 the Seller shall not be liable to the Buyer for any indirect or consequential loss or damage (whether for loss of profit, loss of business, depletion of goodwill or otherwise), costs, expenses or other claims for consequential compensation whatsoever which arise out of or in connection with the Contract."

42. As I will explain later, in light of the documents I was shown and the evidence I heard, I am satisfied that the Stour Bay T&Cs also appeared on the reverse of all or almost all of the invoices rendered by Stour Bay in respect of sales of the Vitamin D3 500 Feed Grade product and other products to Provimi prior to January 2015 when the sales which are the subject of the present action were made.

(iii) FBL's Vitamin D3 500 Feed Grade Product

43. Historically, the global market for vitamin D3 for use in the animal feed industry has been dominated by Chinese manufacturers. Mr Gibbons' evidence was that, prior to FBL's entry into the market, there was only one other non-Chinese company supplying vitamin D3 to the animal feed industry, DSM Nutritional Products AG.
44. At some point during 2009 Mr Gibbons learnt from another Stour Bay customer that FBL was producing a vitamin D3 product and that it was interested in supplying it to the animal feed market in Europe. On 2 October 2009 Mr Gibbons contacted FBL by email. On 5 October 2009 he received a reply from FBL indicating that it produced three vitamin D3 products:

i) Vitamin D3 40 MIU USP/EP for pharma applications;

ii) Vitamin D3 15 MIU in Soya bean oil for feed applications; and

iii) Vitamin D3 100 CWD for Pharma and Food applications.

45. Mr Gibbons enquired the same day whether FBL could also produce vitamin D3 500 (a product with a specification potency of a minimum of 500,000 IU/g of vitamin D3, equivalent to 12.5mg vitamin D3 per gram), and he was told that it could. Mr Gibbons asked FBL for samples of its vitamin D3 500 product, and these arrived at Stour Bay a few days later.
46. On 14 October 2009 Mr Gibbons and Ms Jackson met with Ramya Gurijala of FBL at the CPhl conference in Madrid. In March 2010 Mr Gibbons and Ms Jackson visited FBL at its premises in India, and they discussed with Syamal Ram Kishore, FBL's President of Business Development, and his colleagues the possibility of supplying its vitamin D3 500 product in Europe.

(iv) Applicable EU Regulations

47. At the time the relevant contracts of sale were concluded, the regulation of feed additives, including vitamin D3 500 feed products, was the subject of Regulation (EC) No. 1831/2003.
48. Under Article 3 of Regulation 1831/2003, no person was permitted to place on the market, process or use in the EU a feed additive unless it was covered by an authorisation granted in accordance with the regulation. Article 7 dealt with the authorisation process. Article 10, however, provided a time-limited derogation for additives placed on the market under a former regulatory regime:

"1. By way of derogation from Article 3, a feed additive which has been placed on the market pursuant to Directive 70/524/EEC … may be placed on the market and used in accordance with the conditions specified in Directives 70/524/EEC … and their implementing measures …, provided that the following conditions are met:

(a) within one year of the entry into force of this Regulation, persons first placing the feed additive on the market or any other interested parties shall notify this fact to the Commission. At the same time, the particulars mentioned in Article 7(3)(a), (b) and (c) shall be directly sent to the [European Food Safety Authority];

(b) within one year of the notification mentioned under (a), the [European Food Safety Authority] shall, after verification that all the information required has been submitted, notify the Commission that it has received the information required under this Article. The products concerned shall be entered in the Register. Each entry in the Register shall mention the date on which the product concerned was first entered in the Register and, where applicable, the expiry date of the existing authorisation.

(2) An application shall be submitted in accordance with Article 7, at the latest one year before the expiry date of the authorisation given pursuant to Directive 70/524/EEC for additives with a limited authorisation period, and within a maximum of seven years after the entry into force of this Regulation for additives authorised without a time limit …"

49. Vitamin D in the form of vitamin D3 (its chemical name is cholecalciferol) had been authorised without time limit under Directive 70/524/EEC for use in, inter alia, chickens and poultry, and the additive was entered into the Register of Feed Additives as an existing product in accordance with Article 10(1) of Regulation (EC) No. 1831/2003.
50. In accordance with Article 10(2), applications were subsequently made to the European Food Safety Authority by three manufacturers, including FBL, in 2012, 2013 and 2014, for re-evaluation of vitamin D3 as a feed additive, and in Commission Implementing Regulation (EU) 2017/1492 its approval for use as a compound - not as a specific product produced by a specific manufacturer - was confirmed.
51. So far as FBL's own Vitamin D3 500 Feed Grade product is concerned, the position when Mr Gibbons first spoke to FBL in 2009 was that the product had not yet been accredited for use in Europe. At some point during the second half of 2010, however, with some assistance from Stour Bay, FBL obtained FAMI-QS certification, FAMI-QS being one of the three European accreditation schemes for specialty feed ingredients.

(v) Initial communications with Provimi

52. On 23 April 2010, after his meeting with FBL in India but before the Vitamin D3 500 Feed Grade product obtained FAMI-QS accreditation, Mr Gibbons sent an email to Ken Nakatani, a consultant acting on Stour Bay's behalf in connection with its dealings with FBL, asking if FBL could send a 1 kilogramme sample of the product for distribution to Stour Bay's customers.
53. One of Stour Bay's customers was Provimi, and towards the end of April 2010 Mr Gibbons contacted Provimi's central purchasing department in the Netherlands to enquire whether Provimi might be interested in purchasing vitamin D3 of Indian origin. On 29 April 2010 Mr Piccolin responded by email in the following terms:

"recently you were in contact with Joanna about vit D3, Indian origin. We know the origin by name but did not approve it till now.

You mention it's a D3 water soluble. On our side we consider all D3 are water soluble. But do you know if it's a gelatin free or not?

For us it could be an opportunity sometimes, but it means that we will have first to approve the goods. It means receive a sample, all certificates etc … (you know our procedure).

Consequently, first is to make clear what kind of D3 it is, if you position it as a specific grade or a standard one and thereafter if you agree we will send you all details for starting approval"

(emphasis added).

54. The third sentence of the second quoted paragraph that I have emphasised is of some significance given Provimi's case that the Provimi Gelatin Specification was incorporated into the contracts of sale, and that, as a result, Stour Bay contracted to sell to Provimi a Vitamin D3 500 Feed Grade product that contained a coating of non-ruminant gelatin.
55. So far as this is concerned, the experts agreed that the typical form of vitamin D3 used for supplementing livestock feed comprises spray-dried beadlets, usually prepared using a matrix such as gelatin within which the vitamin D3 is disbursed. As Mr Piccolin explained in his witness statement, however, gelatin-free products are demanded by some customers, mainly those in Muslim countries.
56. As it happened, Stour Bay had received a similar question from another potential customer for the Vitamin D3 500 Feed Grade product only a few days earlier. On 23 April 2010 the customer (whose name was redacted in the email) had asked:

"Dear Nick, can you still offer Vitamin D3 500.00? But we need the following quality: … that is incorporated in a matrix of gelatin and sugar, stabilized with antioxidants."

Mr Gibbons passed the question on to FBL, who replied the same day saying:

"We regret to inform you that we are not offering Vitamin D3 500 feed grade incorporated in a matrix of gelatin and sugar."

This response was then passed on to the customer, who then said that it could not purchase the material.

57. When Mr Gibbons received Mr Piccolin's 29 April 2010 question, "But do you know if it's a gelatin free or not?" he passed it on to Mr Nakatani. Mr Nakatani responded, saying that he had forwarded the message to FBL which had confirmed that its Vitamin D3 500 Feed Grade product had no matrix of gelatin and sugar. Mr Gibbons then responded to Mr Piccolin on 30 April 2010:

"I am advised it is Gelatin free.

If you would like a sample where would you like us to send it?"

58. Unlike the other potential customer, Mr Piccolin's response was not to reject the product on the ground it did not contain gelatin – he had asked if it was gelatin free, and he had been told in terms that it was - but to ask Mr Gibbons to send Provimi a sample of the product and various pieces of product information and documentation to Mr van der Eijk who was copied on the email.

(vi) The Provimi approval process

59. As explained by Mr van der Eijk in his witness statement, before any product can be purchased by a Provimi company, a two-stage approval process has to be undergone:

i) There has to be an evaluation and approval of the product itself; and

ii) There also needs to be approval, if necessary following an audit, of the proposed supplier.

60. The details of Provimi's approval processes at the time are set out in a flow chart and also in a Provimi document that was put to Mr van der Eijk in cross-examination headed "Selection of Supplier and Ingredient" ("the Approval Process Document"). The document makes clear that the processes described in it have to be followed for what is termed a "New SIC", meaning where a supplier is new, where the ingredient is new, or where both are new.
61. The procedure to be followed for the material evaluation is set out in paragraph 6.2 of the Approval Process Document. This provides (in summary) that:

i) Provimi's Quality Assurance department receives a sample of, and documents relating to, the product and checks if the sample is correct and the documentation is complete. If the documentation is complete it is uploaded into Provimi's system;

ii) Quality Assurance then:

"… checks if a PAC number is already assigned and a Provimi specification is available".

Mr van der Eijk explained in cross-examination that the Quality Assurance department needs to have a specification or a PAC number to know what it was checking for;

iii) Quality Assurance then arranges for the sample to be analysed in the chosen laboratory "for all parameters mentioned in the Provimi specification". The results of the analysis are then uploaded into Provimi's system; and

iv) Quality Assurance compares the results of the analysis with the specification. If all the parameters are within tolerances, the approval process will continue; if one or more parameters are out of tolerance, the approval process will be blocked pending further discussions.

62. I should say something about the PAC Code referred to in paragraph 61 ii) above. Mr Piccolin was asked in cross-examination what the acronym "PAC" stood for. He had, in fact, answered that specific question in his witness statement – it stands for "Provimi Article Code" – but in his oral evidence he explained what the code meant in practical terms, how it was used, and how it related to the approval process:

"Q. … The PAC number, what does 'PAC; stand for at the top of –

A. It is a characteristic of the product we are using, vitamins or not vitamins. You have a specific PAC code for each product we are using.

Q. I see. So what would happen is that someone from the particular plant, let's use Provimi France, says, 'I want the product that …' – it's almost like an administrative number. So if he says, 'I want some product PAC …' – I can't remember the whole of the number but it ends in a '1' at the end – then you know what product you have to, as a buyer, go out and buy?

A. Yes.

Q. But again, as we've said, it's not a case of finding any old product. It's finding an approved product which the quality control people have told you fits those requirements?

A. Yes."

63. The supplier approval process was addressed in paragraph 6.3 of the Approval Process Document. This provided for the completion and submission by the proposed supplier of a questionnaire with supporting documents. Assuming all the questions were answered and all the necessary documents were supplied, the results would be uploaded into Provimi's system and would then be given a numerical score. Depending upon the total score, the supplier was classified as low risk (green), medium risk (amber) or high risk (red). A low risk classification meant that the SIC could be approved without an audit, unless an audit was required for other reasons; in the case of medium risk classification, an audit was optional; a high risk classification would result in the SIC being rejected.
64. Mr Piccolin's evidence was that, as a global buyer for Provimi, he was not able to place an order for a product until that product had been expressly approved by Provimi's Quality Assurance department. Mr Amouroux's evidence was to the same effect. Their evidence was consistent with paragraph 6.3.11 of the Approval Process Document, which provided, in the case of a rejected SIC, that:

"SIC stays blocked. Purchasers are not allowed to buy this material from this supplier".

(vii) Direct communications between FBL and Provimi

65. Although Mr Piccolin asked Mr Gibbons on 30 April 2010 for a sample of FBL's Vitamin D3 500 Feed Grade product (which Mr Gibbons had confirmed was gelatin-free) to be sent to Mr van der Eijk, as explained earlier the product was not, at that stage, accredited for use in Europe.
66. On 10 May 2011 FBL approached Mr Piccolin directly. An email from Devendra Choubey, Provimi's Deputy Manager – Business Development explained that FBL had a range of four vitamin D3 products, and that it was already supplying two of them – including its Vitamin D3 500 Feed Grade product - to Provimi India. FBL said that it was keen to supply Provimi businesses in Vietnam and Russia. The four FBL products identified were:

i) Vitamin D3 40 MIU USP/EP for Pharma applications;

ii) Vitamin D3 2/4/5/15 MIU in soya bean Oil for feed applications;

iii) Vitamin D3 500 Feed grade for feed applications; and

iv) Vitamin D3 1MIU in corn Oil USP for Pharma and food applications.

67. Mr Piccolin responded by email on 12 May 2011 explaining that Provimi needed to approve FBL's products and asking Mr Choubey to send samples for all products for which it sought approval to Andy van Waveren of Provimi in the Netherlands. Mr Piccolin explained in cross-examination that he was aware that Provimi India sourced some products locally, although he said he was not aware that it was dealing with FBL.
68. Mr Choubey replied on 13 May 2011 asking Mr Piccolin – seemingly because FBL produced four different strengths of vitamin D3 oil - if Provimi could provide him with the specifications for the vitamin D3 oil that it was currently procuring. Mr Piccolin responded on the same day attaching four different specifications. Two were, as asked, for vitamin D3 oil, but the other two were for different forms of vitamin D3 500.
69. The first of these specifications, identified in the list of attachments to the email as "Vitamin D3 500.pdf", was Provimi Ingredient Specification, PAC number 191910101, i.e., the Provimi Gelatin Specification referred to in paragraph 12 (i) above.
70. The first heading in this document was "Ingredient" and the first item dealt with under that heading was "Product Information", which read as follows:

"Cholecalciferol (or Vitamin D3) mixed with a cereal-derived carbohydrate carrier and a coating of non-ruminant gelatin."

This language was the basis of Provimi's case that, by contracting on terms which included the Provimi Gelatin Specification, Stour Bay had contracted to sell to Provimi a Vitamin D3 500 Feed Grade product that was coated with non-ruminant gelatin.

71. The second vitamin D3 500 specification sent by Mr Piccolin was referred to in the list of attachments as "Vitamin D3 500 GELATIN FREE.pdf". This was a Provimi Ingredient Specification, PAC number 191910102 (so, one number different from that of the Provimi Gelatin Specification) for Vitamin D3 500 Gelatin Free ("the "Provimi Gelatin Free Specification").^[1] The Product Information under the first heading in this specification stated that the ingredient was:

"Cholecalciferol (or Vitamin D3), finely dispersed, spray-dried in a matrix of vegetable protein and dextrin"

72. It is not clear exactly when this was provided to Provimi, but on 2 June 2011 FBL completed a Provimi information document that contained two sections, a section headed "Manufacturers Company Information", and another headed "Supplier Questionnaire for Manufacturers". As I explain later, a somewhat similar document was completed later by FBL in 2012 as part of the supplier approval process.
73. The Supplier Questionnaire section included the following question:

"2.9) Do you handle materials that contain any form of animal proteins (or their derivatives), fish products or fish meal?"

FBL answered the question in the negative, thus confirming that it did not handle materials containing any form of animal protein. On the basis that gelatin is a form of animal protein, Stour Bay says that the answer confirmed what Mr Piccolin had already been told in April 2010, namely that FBL's Vitamin D3 500 Feed Grade product did not have a gelatin coating.

(viii) The Material Evaluation and Supplier Audit

74. On 3 June 2011 Mr Gibbons learnt from Mr Nakatani that FBL had been in direct contact with Provimi and that Provimi had requested product samples for approval, which had been sent.
75. Mr Gibbons then followed up on his own discussions with Mr Piccolin. On 20 June 2011, responding to Mr Piccolin's email of 30 April 2010 (in the same email chain as the email in which Mr Piccolin had been told that FBL's product was gelatin free), Mr Gibbons said that FBL had now received FAMI-QS accreditation. He asked Mr Piccolin whether the required information should still be sent to Mr van der Eijk.
76. Mr Piccolin did not immediately respond, and so on 6 July 2011 Mr Gibbons sent an email directly to Mr van der Eijk. Mr Gibbons explained that Mr Piccolin had given him Mr van der Eijk's name some time ago, and he attached many of the documents that Mr Piccolin had previously indicated were required, telling Mr van der Eijk that he would forward a sample of FBL's Vitamin D3 500 Feed Grade product.
77. Mr Piccolin eventually responded on 11 July 2011 explaining that Provimi's procedures were unchanged but that sampling approval process was now under the umbrella of Mr van Waveren, and so the necessary samples and documentation should be sent to him. Having listed the samples and documents required, Mr Piccolin explained that:

"The status of Stour Bay within our system is approved with conditions, meaning the origin of the product has to be from approved supplier. For instance, in this case, it means the producer of D3 has to be approved, thus filling up the following questionnaire:

…

As I don't know exactly where do we stand with this D3, I prefer to restart from scratch, but maybe you had already some exchanges with Laure or Andy. In this case just follow it up."

78. Later on 11 July 2011 Mr Piccolin sent a further email to Mr Gibbons, explaining that he had seen Mr Gibbons' email to Mr van der Eijk, that FBL had contacted Provimi directly and that an approval process was already underway:

"Fermenta has been contacted us and we have started an approval procedure with them! I suppose they are trying different channel.

We are awaiting a sample and have sent a questionnaire too.

I suggest we wait for final approval and then we will be able to buy from Stour Bay in case you provide good price and good service.

I prefer to be transparent rather than not replying anything to you."

79. Mr Gibbons responded to Mr Piccolin on 12 July 2011 saying that he was aware that FBL was talking directly to Provimi and that FBL had asked Stour Bay to get involved. Mr Gibbons said that he had samples and the relevant forms. Emails from Mr Gibbons to Mr van Waveren show that they were sent to Provimi either on the same or on the following day.

(a) The material evaluation

80. As I explained earlier, the Provimi approval process has (or had at the relevant time) two components: evaluation of the product (or material), and approval (if appropriate, following an audit) of the proposed supplier. The supplier approval process, as Mr Piccolin had said in his 11 July 2011 email, meant that it was necessary for FBL as well as Stour Bay to be approved.
81. So far as the material evaluation is concerned, as Mr van der Eijk explained in his evidence Provimi was provided with samples of various FBL products for testing including its Vitamin D3 500 Feed Grade product. The email correspondence suggests that the samples were despatched in late June 2011, although it appears they were not analysed by Provimi until August 2011.
82. Two Provimi laboratory reports appear in the bundles (bearing sample numbers 213355 and 213356), both dated 31 August 2011 and signed by Mr van Waveren. They report on moisture levels, the quantities of certain heavy metals in the samples (including mercury), particle size distribution, and the quantity of vitamin D3. In both reports the vitamin D3 level was shown as in excess of 500,000 IE/g.
83. As explained in paragraph 61 iii) above, paragraph 6.2 of the Approval Process Document required Provimi's Quality Assurance department to analyse the sample for "all parameters mentioned in the Provimi specification". Mr van der Eijk was asked in this context about the fact that the Provimi Gelatin Specification identified the ingredient as vitamin D3 with "a coating of non-ruminant gelatin":

"Q. So the non-ruminant gelatin coating was a key part of this product, wasn't it?

A. Yes, you can consider that.

Q. It was an important parameter?

A. No, not necessarily parameter, because parameters are listed under section 2, 3 and 4 [of the Provimi Gelatin Specification].

Q. Well, surely a coating which you, I'm assuming, as a material specialist, would agree is vital when you have particular ingredients mixing with the vitamin D supplement, the coating is an important, isn't it, integral part of the product?

A. Yes.

Q. So you're never going to know, are you, whether the product is going to work unless you check that the coating is intact?

A. I would say yes and no. Part of this as well is how far -- what are the analyses that we can do on the product and how much do we rely on the supplier to have feed knowledge and premix knowledge. In principle, from my perspective, this is not a parameter because it's not listed as a to-be-defined test to be done, to be determined if the coating is non-ruminant gelatin or not. In principle, the exact, let's say, coating itself does not really matter. I think the more important here is that: is the product coated, yes or no; instead of what is the origin of the coating.

Q. Right. So you've got to decide whether the product is coated, yes or no, and, as I understand it –

A. Yes."

84. As can be seen from this extract, whilst Mr van de Eijk did not regard the type of coating as of particular significance (although, in the case of a product to be used in a Muslim country, one would think it might be), he accepted that Provimi had to determine whether the sample of product was coated or not. As he later explained, however, no testing was carried out by Provimi to see if the sample was coated:

"Q. But isn't one of the things you've got to test for, as I understand it, stability, the stability of vitamin D3 is known to be impacted by aggressive ingredients, so whether it's a choline chloride or whether it's hydrated iron salts or any of the other items. And having a coating is very important, isn't it, as I understand it?

A. Yes.

Q. So isn't that something that you should be checking? If you are in quality assurance, you're checking that you've got a product that works, shouldn't you be at least testing for the coating?

A. There are different ways of testing and clarifying if the product is coated yes and no. Testing is just one of them.

Q. Right, so what did you do to establish whether or not there was a coating on this product as required by the specification?

A. In this case we didn't test for stability or coating and apparently -- I cannot recall but there was not need for it because we were relying on the supplier to provide us a material according to the specification.

Q. That rather depends, doesn't it, on, one, the supplier having actually been provided with the specification and told to comply with it? You would agree with that?

A. Yes.

Q. And surely your job in quality assurance is to check that someone has complied with the specification; otherwise what is the point of you carrying out any testing at all?

…

Q. We know that one or two -- it's right there, we looked at it in the product specification, that this product was meant to, apparently, have a gelatin coating. Are you saying --

A. Hmm-hmm, correct.

Q. -- you carried out no testing at all to determine whether or not it had a coating?

A. Correct.

Q. Even though you know that not having a coating would, if you like, undermine or could undermine the performance of the product?

A. Correct."

There was no dispute between the experts that, as a matter of fact, FBL's Vitamin D3 500 Feed Grade product was not coated at all.

85. As part of his material evaluation, Mr van der Eijk filled in an Ingredient Risk Assessment form. This was a template form with certain parts pre-filled and with others to be completed. At the top of the first page it identified the Ingredient Name as Vitamin D3 500, the PAC Assignment as 191910101 and the Supplier Name as FBL. Below that was a section which identified the Ingredient Information from the Supplier.
86. Six types of documents were listed, two of which were identified as "Required before completing the risk assessment". These were:

i) MSDS for ingredient from supplier; and

ii) Supplier's ingredient specification.

The first of these was checked by Mr van der Eijk to indicate that it had been provided; the second was not.

87. So far as the first of these two documents is concerned, the MSDS document for FBL's Vitamin D3 500 Feed Grade product, which Mr van der Eijk confirmed in his evidence he had seen, included the following information and warnings:

"7. HANDLING and STORAGE:

…

Storage:

Storage conditions: Store at cool and dry place and Protect from Light.

Validity:

12 months in the unopened original container, after opening the container the contents should be used within a short period.

…

10. STABILITY AND REACTIVITY:

Conditions to Avoid : Heating, Light, Humidity, Air, Warming

Incompatibilities with

Other Materials :

Copper and copper alloys

Iron and iron salts

Strong Oxidizing agents

Peroxides

Strong inorganic Acids

Strong bases"

88. Asked about the "Conditions to Avoid", Mr van der Eijk acknowledged that, if choline chloride was added to the Vitamin D3 500 Feed Grade product in making Provimi's pre-mixture, then, as choline chloride attracts water, humidity could be an issue. He also accepted, having regard to the listed "Incompatibilities", that Provimi's pre-mixture contained hydrated iron and copper salts.
89. When it was put to Mr van der Eijk, however, that these matters suggested that the FBL product was not particularly compatible with Provimi's pre-mixtures his answer was that this was not necessarily so. He said that the safety data contained in the MSDS was included from the perspective of workers' safety, not necessarily for product safety, and that was how it was viewed for the purposes of the material evaluation.
90. Mr van der Eijk was asked about the Suppliers' ingredient specification, the second document that the Ingredient Risk Assessment form said was required to be received. Mr van der Eijk accepted that, because he had not checked the relevant box, he most probably did not have it, and that in the circumstances it had not really been appropriate for him to complete his risk assessment. Any ingredient specification for FBL's Vitamin D3 500 Feed Grade product would not, of course, have included gelatin.
91. The remainder of the Ingredient Risk Assessment form dealt with Ingredient Classification, in this case checked to show that the ingredient was a vitamin, the production process, and storage or packaging. There was then, under the heading "Potential Risk Assessment", a list of safety contaminants which were to be marked by reference to risk impact, risk potential and risk score. The form was completed by Mr van der Eijk to show a total risk score of 14, which put the assessment in the high risk category. An explanatory note said that:

"Due to Hg found in the sample send by supplier the risk for heavy metals is a high potential".

92. On 2 November 2011 Mr Piccolin sent an email to Mr Gibbons explaining that the sample of FBL's Vitamin D3 500 Feed Grade product had been rejected because of the high mercury content. Mr Gibbons responded, explaining that he was aware of a small number of batches of the product that had shown traces of mercury and that the source of the mercury – a starch used in the manufacturing process – had been identified and was being changed.
93. On 3 November 2011 Mr Piccolin contacted FBL directly by email, copying Mr Gibbons, explaining that its approval status was still pending. He explained that a site audit had to be carried out, and in the meantime he asked for another product sample plus a completed manufacturers' questionnaire to be sent, this time not to Mr van der Eijk but to Jan Derk Brouwer at Provimi Netherlands.
94. Provimi did not disclose any details of the further sample sent to Mr Brouwer or any documents concerning Mr Brouwer's analysis of it. Mr van der Eijk accepted that there must have been another approval process, but he had no explanation as to why the documents had not been located. It seems unlikely that all the relevant documents were deleted as a result of the document retention process that I described earlier, which, as I understood it, affected Outlook emails, but no explanation was provided by Provimi as to why these documents were absent. Mr Brouwer was not called as a witness by Provimi to explain the evaluation he had done.
95. At all events, on 2 December 2011, following the completion of Mr Brouwer's approval process, Laure Durand, one of Mr Piccolin's colleagues, sent an email to Mr Gibbons, copied to Mr Piccolin, saying that samples of various FBL products, including its Vitamin D3 500 Feed Grade product, had now been approved, and that a supplier audit of FBL was now needed.

(b) The supplier audit

96. The supplier audit of FBL – the second part of the Provimi approval process – was carried out in October 2012.
97. A Provimi Supplier Questionnaire for Manufacturers was completed by FBL for its vitamin D3 500 and oil products on 3 October 2012. This was somewhat similar to the Manufacturers Company Information and Supplier Questionnaire for Manufacturers document completed by FBL on 2 June 2011 that I described earlier. The document was scored by the two on-site auditors, Mr Shekhar and Mr Brouwer.
98. Among the questions included in the questionnaire was question 17 which asked whether FBL used or handled on its site any of a number of listed materials, including:

"Materials that contain any form of animal proteins (or their derivatives), fish products or fish meal".

The question was answered "none", i.e., that FBL neither used nor handled any of these materials which included animal proteins.

99. Mr van der Eijk did not agree that this answer indicated that FBL could not be producing a product with a gelatin coating; he said that he regarded the question as concerning cross-contamination. His answer was, however, somewhat undermined by the presence of, and the answer to, question 21, which was specifically concerned with contamination. Question 21 asked:

"Do you have controls in place to avoid contamination of your product with animal protein?"

The answer given was "not applicable", the explanation given in the "Comments" box "No animal proteins".

100. Question 55 asked what checks were performed on raw materials and processing aids, requiring FBL to list all the raw materials used and any checks carried out in relation to them. Mr van der Eijk was asked about the absence of any reference in the list of raw materials to gelatin:

"Q. Right, and there's no gelatin in that list?

A. Correct.

Q. Right. So it must have been obvious from that, if nothing else, if you won't agree with me on anything else, it must have been obvious from that list that there was no gelatin coating in this product?

A. Correct."

101. Mr van der Eijk was not dealing with the supplier approval process in 2012; it was being conducted by Mr Brouwer and Mr Shekhar. In the absence of any evidence from either of them, however – and I was given no explanation as to why they were not called as witnesses - I infer that they must have appreciated what Mr van der Eijk himself agreed was obvious from this document, namely that FBL's Vitamin D3 500 Feed Grade product had no gelatin coating.
102. This inference is consistent with the agreement between the experts, recorded in paragraph 17 of their Joint Memorandum:

"Provimi/Cargill were in receipt of information in 2012 (the Provimi Supplier Questionnaire for Manufacturers) that showed that the FBL product was not spray-dried and did not contain gelatine."

Mr Pickard was asked in cross-examination whether, as a result of its examination of the sample and its request for documents, Provimi should have determined that there was no gelatin or vegetable coating. His answer was:

"At that point in time, with that particular product and information at that time, they would have – they should have known, certainly."

As I explained above, I am satisfied that Provimi must have appreciated (and did appreciate) that there was certainly no coating of gelatin on the product, and I infer that it also knew – or certainly should have known – that the product had no coating at all.

103. The conclusion of the supplier audit, recorded at the foot of the questionnaire, was that FBL was approved, the auditors noting that FBL was the only vitamin D3 producer in India and that it produced one tonne of feed grade 500 powder per day. Mr Gibbons was notified of the approval of FBL by an email from Mr Piccolin on 3 December 2012, and FBL by a letter from Mr Brouwer dated 7 December 2012.

(ix) Orders in 2013 and 2014

104. FBL and its Vitamin D3 500 Feed Grade product now having been approved, orders for the product could be placed by Provimi. The first order was placed on 18 February 2013.

(a) The ordering process

105. Before dealing with the detail of the orders, it is necessary for me to say something about the ordering process and about the documents generated by both parties in that regard. All the orders placed by Provimi followed essentially the same pattern.
106. The process is described in paragraph 9 of the Particulars of Claim, paragraph 8 of the Defence and paragraph 2 of the Reply. As paragraph 2(1) of the Reply records, the contractual scheme, and the documentation generated, was – on the pleadings at least – substantially common ground. As I will explain later, Mr Kulkarni, QC sought to move away from that common ground somewhat in his submissions.
107. Provimi operates (or operated at the relevant time) a centralised purchasing process whereby each of the individual Provimi companies in different countries provides a forecast of its requirements for various ingredients. These are given to Provimi's central purchasing team in the Netherlands, which then consolidates requirements for the same product and places bulk orders with third-party suppliers.
108. In the case of the purchases of FBL's Vitamin D3 500 Feed Grade product from Stour Bay the starting point was always a verbal (or email) negotiation and agreement on price, quantity and delivery period between Mr Piccolin and/or Mr Amouroux of Provimi and Mr Gibbons of Stour Bay.
109. So far as the first purchase is concerned this took place in an exchange of emails on 15 and 18 February 2013:

i) On 15 February 2013 (seemingly after meeting for them lunch) Mr Gibbons emailed Mr Piccolin and Mr Amouroux suggesting that, for a trial run of the product, Provimi might want to order four pallets;

ii) Mr Piccolin responded on the same day suggesting a trial with a small volume – he suggested 3 pallets or 900kg - for Provimi France for delivery in May or June 2013. Mr Piccolin added:

"We can do the contract now, but the price will have to be revised just before delivery of course. In order to issue our contract, what would be your best offer for a spot delivery? We will start from there";

iii) Mr Gibbons responded on 18 February 2013 agreeing to a 900kg trial and asking whether a price of EUR 6.00/kg delivered was agreeable. He said that the stock was ready – Stour Bay would, it seems, frequently purchase and hold stock of FBL's Vitamin D3 500 Feed Grade product in advance of orders being received;

iv) On 18 February 2013 Mr Piccolin replied to Mr Gibbons' email saying:

"As discussed contract has been made. As explained, we don't know yet when call off will be done. It looks like it will be for Q2, but we leave the option to our location.

The price of 6 € is ok today. In case market decrease, we will ask you to adjust of course, as agreed. At least there is something in place as a start."

110. Once these basic terms had been agreed, Mr Piccolin or Mr Amouroux would then instigate an automated or semi-automated process of generating and sending Stour Bay a written contract confirmation. The document was headed "PARS Agreement" (Mr Piccolin explained that the acronym stood for "Provimi Agreement Registration System") and was what the parties referred to as a "Parent PARS Agreement".
111. Thus, in the case of the trial purchase agreed in the emails between Mr Piccolin and Mr Gibbons on 18 February 2013, at 15:06 on 18 February 2013 an email was sent from a central Provimi email address support@gep.com to Mr Gibbons, copied to Ms Pickford, in the following terms:

"Dear Supplier,

This is to notify you that the 2013000234 – Vitamins Contract has been executed by Pierre Piccolin on 2/18/2013 10:03:31 AM.

Attached you can find our Agreement containing terms and conditions for the coming deliveries of your products to our affiliates. Individual Purchase Orders and delivery details will be issued by local Cargill entities, in accordance with this agreement.

It is imperative that the Agreement number is used as a reference on all correspondence with Cargill Animal Nutrition affiliates, including shipping documents, delivery notes and invoices. This will ensure trouble free receipt of your products and prompt payment.

If there are any immediate concerns regarding the Agreement please contact Pierre Piccolin.

Cargill Animal Nutrition, Risk Management & Sourcing".

112. Attached to this email was the Parent PARS Agreement and the two attachments referred to in it.
113. The Parent PARS Agreement itself was a two-page document:

i) The document had the header "PARS Agreement – 2013000234" and underneath that the main heading "Contract Information";

ii) There was then a sub-heading "Agreement Information" under which was listed the Agreement number - 2013000234, the Product category – Vitamins, and the Agreement Date – 2/18/2013. There followed a sub-heading "Supplier details" which identified Stour Bay as the supplier and gave contact details;

iii) The next sub-heading was "Details". Beneath that, the document listed:

a) The Product Name: Vitamin D3 500;

b) The PAC code: 191910101;

c) The Price per Unit: EUR 6.00;

d) The Quantity: 900kg; and

e) The Location/Profit Centre: Provimi France - Crevin.

There were then five items A to E. Item A referred to Inco Terms 2010 and DDP (delivered duty paid) and item E referred to the Producer, which was identified as "origin Fermenta, India";

iv) Then there was a sub-heading "General Terms" under which was the following:

"Provimi standard minimal payment terms are 60 days from end of month following deliveries. These terms should not override any payment term which exceed 60 days from end of month following delivery already agreed and in place locally. For this specific agreement, the agreed payment terms are stated above.

Formal purchase orders and delivery details will be confirmed by the local Provimi companies.

…

Start Delivery Period (mm/dd/yyyy) 2/18/2013

End Delivery Period (mm/dd/yyyy) 6/30/2013";

v) On the second page of the document was a sub-heading "Remarks". Most of the remarks concerned delivery and call off arrangements, but they included a "baisse clause", reflecting the agreement between Mr Piccolin and Mr Gibson in their email exchanges about a possible adjustment of the price:

"baisse clause: in case market prices fall down, seller will adjust the contract price down, to match the market price. In case, there is no agreement between Stour Bay/Fermenta and Cargill/Provimi contract might be cancelled";

vi) Below that, the origin of the product was again identified as FBL, India;

vii) There was then a sub-heading "Contract Attachments", which identified two documents that were attached to the Parent PARS Agreement and the email.

114. The first attachment, identified in the Parent PARS as "Vitamin D3 500.pdf", was the Provimi Gelatin Specification, PAC Code 191910101.
115. As I explained earlier, a copy of the Provimi Gelatin Specification had been sent by Provimi to FBL, along with the Provimi Gelatin Free Specification and two specifications for vitamin D3 oil products, in May 2011.
116. There was no evidence, however, and it was not suggested by Mr Kulkarni, QC, that the Provimi Gelatin Specification had ever previously been sent to, or seen by, Stour Bay. Nor was there any evidence that the terms and the intended application of the Provimi Gelatin Specification had ever been discussed, let alone agreed, between Mr Piccolin and/or Mr Amouroux and Mr Gibbons, either in relation to the trial order on 18 February 2013 or in relation to any subsequent order. So far as Stour Bay is concerned, the appearance of the specification as an attachment to the Parent PARS Agreement was, therefore, unheralded by any prior communication.
117. In paragraph 12 of the Particulars of Claim Provimi alleged that:

"In or about 2012, the Claimants invited the Defendant to quote for the provision of a Vitamin D3 product ('the Product') and, as part of that invitation to quote, the Claimants supplied the defendant with a product specification ('the Specification'). The Specification required:

(1) that the Product should contain 'a coating of non-ruminant gelatin'".

118. The allegation in this paragraph, that Provimi had provided Stour Bay in 2012, as part of an invitation to quote, with a copy of the Provimi Gelatin Specification, was not pursued by Mr Kulkarni, QC and is simply wrong. There is no evidence that the Provimi Gelatin Specification was seen by Stour Bay at any time prior to its receipt as an attachment to the first Parent PARS Agreement on 18 February 2013.
119. The second attachment to the Parent PARS Agreement was an FBL product information document headed "Vitamin D₃500 Feed Grade. I will refer to this as "the FBL Product Information Sheet". This provided some particulars of FBL's Vitamin D3 500 Feed Grade product, describing the product as an "Animal feed supplement", and it included an image of the product bag and label, which contained the following instructions:

"store in cool place below 25.00 C. Use immediately after opening."

120. The Product Information Sheet identified the Assay of the product as:

"between 0.48 to 0.55 MIU/g of Vitamin D3 content."

which was different to the vitamin D3 content shown on the Provimi Gelatin Specification, which was a minimum (with no specified maximum) of 500,000 IU/g. The Provimi Gelatin Specification and the FBL Product Information Sheet were, to that extent, inconsistent.

121. In the case of the 18 February 2013 trial order, the relatively modest quantity of Vitamin D3 500 Feed Grade product purchased was delivered to a single company, Provimi France. Subsequent purchases of larger quantities of product made by Provimi's central purchasing arm, however, were divided up between different Provimi companies. Where this happened, a number of additional documents were generated.
122. The first of these was what the parties referred to as a "Child PARS Agreement". The second order on 12 July 2013 was also for a quantity of Vitamin D3 500 Feed Grade product that was delivered solely to Provimi France (albeit in a number of instalments), but the third order on 19 November 2013 for a quantity of 5,000kg of product was ultimately split amongst a number of Provimi companies.
123. The documentation in relation to this third order included the following:

i) A Parent PARS Agreement, number 2013002468 for the full quantity of (initially) 5,100kg. The Location/Profit Centre on this was identified as "All" and the place of delivery stated simply to be "Europe". The quantity was amended to 5,000kg by the issuance of a replacement Parent PARS Agreement bearing the same number later the same day;

ii) Also on the same day, by emails sent from the same support@gep.com address, other documents also headed PARS Agreements were issued by Provimi's central purchasing team, but this time for smaller quantities to be delivered to particular Provimi companies in particular locations and within particular delivery periods. These were the Child PARS Agreements:

a) Child PARS Agreement, number 2103002483 – initially 2,100kg, later amended to 2,000kg, for Provimi Poland;

b) Child PARS Agreement, number 2013002484 – initially 900kg, later amended to 1,000kg, for Provimi Spain; and

c) Child PARS Agreement, number 2013002485 – initially 1,200kg, later amended to 1,500kg, for Provimi UK.

The 500kg remaining under Parent PARS Agreement, number 2013002468 appears to have been rolled into the next order that was placed in February 2014;

iii) These Child PARS Agreements, aside from the fact that they were for smaller quantities sold to different companies, were generally similar to the Parent PARS Agreement to which they related (which was identified by number on the first page), including a reference to PAC Code 191910101 and attaching the Provimi Gelatin Specification and the FBL Product Information Sheet;^[2]

iv) The Child PARS Agreements were generated centrally, not by the relevant "child" Provimi company, but a copy was sent to that company as well as to Stour Bay. Each Provimi company would then issue its own documents, including in some cases its own internal contract form and in all cases its own written purchase order which was then sent to Stour Bay;

v) So far as the Claimants are concerned:

a) Three of the four Claimants, Provimi France, Provimi Spain and Provimi UK (but not Provimi Poland) issued their own, internal contract documents (a Contrat d'Achat (Provimi France), a Contrato/Contract (Provimi Spain) and a Purchase Contract (Provimi UK)). Save in the case of Provimi Spain, these were not sent to Stour Bay;

b) Each of the Claimants, however, sent its own formal purchase order to Stour Bay (a Commande d'Achat (Provimi France), an Order de Compra/Purchase Order (Provimi Spain), an Order (Provimi Poland) and a Purchase Order (Provimi UK)). Save in the case of Provimi Poland, these gave the number of the relevant Child PARS Agreement;

c) Some of these purchase orders, but not those issued by Provimi Poland, referred to PAC Code 1919101011. This was not the same as the code for the Provimi Gelatin Specification; it had an extra numeral "1" at the end. They all referred to the product as Vitamin D3 500.

The trial bundles included similar documentation in relation to purchases by other Provimi entities in Belgium, Hungary, Ireland and the Netherlands, although in the case of some companies a less formal email call-off appears to have been used in place of a formal purchase order;

vi) Following receipt of a purchase order or call-off from the relevant Provimi company, Stour Bay (Ms Pickford, Stour Bay's Shipping Manager) would usually reply by email or telephone confirming the purchase order and agreeing delivery details, sometimes enclosing a delivery note; but sometimes she would simply arrange for delivery of the product by haulier;

vii) Once delivery of the product to the relevant Provimi company had taken place and proof of delivery had been received, Ms Morgan of Stour Bay would produce an invoice, which would be printed and sent to the company by first class post accompanied by a Certificate of Analysis. The invoices would then be paid by the individual Provimi companies.

124. Provimi's pleaded case as to the legal effect of the arrangements I have described is set out in paragraphs 9 and 11 of its Particulars of Claim:

"9. Thereafter, from or about May 2013, the Claimants began purchasing the Product from the Defendant. They did so as separate entities and individually but with similar contractual arrangements with the Defendant, as explained below.

…

11. On the true and proper analysis of the above contractual arrangements, each supply of the Product by the Defendant to the Claimants under a particular Purchase Order was a separate sale contract under the umbrella of the relevant Child PARS Agreement and, in turn, Parent PARS Agreement."

125. The proposition contained in paragraph 11 of the Particulars of Claim, that the end-result of these arrangements was separate contracts of sale between Stour Bay and the individual Provimi companies, was common ground. In paragraph 8.2 of its Defence, Stour Bay pleaded that:

"8.2 It is admitted, to the extent set out below, and insofar as it is not alleged, averred, that the Claimants, Provimi Belgium, Provimi Hungary, Provimi Ireland and Provimi Netherlands purchased the FBL Product as separate entities and individually but with similar contractual arrangements".

The only substantive pleaded issues between the parties in relation to the contractual arrangements were whether the Provimi Gelatin Specification formed part of the relevant contracts of sale and whether the Stour Bay T&Cs were incorporated into them by a course of dealing.

(b) The early orders

126. The contracts of sale under which the present dispute arises were made in and after January 2015. They were, however, preceded by a number of orders placed in 2013 and 2014; and, as the extent of the prior dealings between Provimi and Stour Bay is relevant to the argument between the parties about the incorporation of the Stour Bay T&Cs, it is necessary to say something about them.
127. As I explained earlier, the first order, which the parties themselves referred to as a trial order, was placed on 18 February 2013. It appears that the order was delivered to Provimi France in late April 2013 and that on 29 April 2013 Stour Bay issued an invoice to Provimi France (invoice 4970) for the 900kg delivered. The invoice had the Stour Bay T&Cs printed on the reverse.
128. There was a rather arid debate between the parties as to whether the next three orders placed in July 2013, November 2013 and in February 2014 were also trial orders. Mr Gibbons described them as such in his witness statement, but they were not referred to in this way in the documents, and in my judgment whether these orders were trial orders or "ordinary" orders is of no legal significance.
129. The orders, which I refer to below by reference to the date and number of the relevant Parent PARS Agreements, were as follows. In some cases, the Parent PARS Agreement was initially issued for a specified quantity, but then amended to a different quantity; the table below identifies only the final agreed quantity:

Date	Parent PARS Agreement	Quantity
12 July 2013	2013001532	7,200kg
19 November 2013	2013002468	5,000kg
17 February 2014	2014000300	4,000kg
17 December 2014	2014001728	22,000kg
		Total: 38,200kg

130. As I indicated earlier, the second order placed on 12 July 2013, like the first, was for a quantity of Vitamin D3 500 Feed Grade product to be delivered solely to Provimi France. The third, fourth and fifth orders, however, were all split, and Child PARS Agreements were accordingly produced followed by purchase orders issued to Stour Bay by the different Provimi companies.
131. As these 2013 and 2014 orders are not the subject of any claim, it is not necessary for me to go into detail about them. A schedule headed "Invoice History" which accompanied Ms Ansell, QC's written opening submissions (substantially replicated in a further schedule handed up to me during her oral opening submissions) lists the invoices rendered by Stour Bay to the four Claimant Provimi companies in respect of these transactions.^[3]
132. The schedule shows that:

i) 11 invoices were rendered to the First Claimant, Provimi France, between April 2013 and March 2015;

ii) Two invoices were rendered to the Second Claimant, Provimi Spain, between February and December 2014;

iii) Six invoices were rendered to the Third Claimant, Provimi Poland, between December 2013 and May 2015; and

iv) 11 invoices were rendered to the Fourth Claimant, Provimi UK, between November 2010 and April 2015.

133. The trial bundles also included invoices rendered during the same period to other Provimi companies both for Vitamin D3 500 Feed Grade and for other products, e.g., invoices 4042, 4032, 4245, 4337 and 4418, all addressed to SCA Nutrition (later Provimi Ireland), for choline chloride, and invoice 6297, addressed to Provimi Ireland, for Vitamin D3 Feed Grade product.

(c) The practice in relation to the Stour Bay T&Cs

134. The evidence of Alison Morgan, who was primarily responsible for preparing and posting invoices to Stour Bay's customers, was that, once the relevant goods had been delivered, she would produce invoices using Sage software which would be printed on special Stour Bay headed paper that had the Stour Bay T&Cs printed on the reverse, a stock of which, she said, was located by the printer that was used for printing invoices.
135. There were, Ms Morgan explained in her cross-examination, rare occasions when invoices would be produced by other Stour Bay employees while she was away. In that event, an invoice might be produced manually, not using the Sage software, but all the other procedures, she said, would be the same. Invoices were checked either by Mr Gibbons or by Ms Jackson before they were sent out.
136. Ms Morgan explained that there was a time in 2015 – she agreed with Ms Jackson that it was a period of around three to five months between April and September 2015 – when, due to an issue with the paper supplier, Stour Bay ran out of the special pre-printed paper. Other than that, however, Ms Morgan said she could not recall any occasion when invoices were not printed with the Stour Bay T&Cs on the reverse.
137. During the period when the pre-printed paper was not available, Ms Morgan explained that she put in place a workaround: she either manually printed the Stour Bay T&Cs on the back of an invoice printed on ordinary Stour Bay headed paper, or she produced a copy of the Stour Bay T&Cs as a separate, stand-alone document which she then included in the envelope with the invoice when it was sent to the customer.
138. It was suggested by Mr Kulkarni, QC to Ms Morgan in cross-examination that the system for attaching Stour Bay's T&Cs was sometimes not followed. Ms Morgan disagreed: she said that it would always have been followed, and that every invoice that was sent out would have been accompanied by the Stour Bay T&Cs, either printed on the back of the invoice or accompanying the invoice as a stand-alone document.
139. Ms Morgan worked two days a week (Monday and Tuesday); her colleague Ms Pickford worked on the other three days. Whilst Ms Morgan was principally responsible for preparing and issuing invoices, Ms Pickford said that she was occasionally required to issue an invoice, and that when she did the invoices would be printed on Stour Bay headed paper with the Stour Bay T&Cs on the reverse.
140. The evidence of Ms Morgan and Ms Pickford about Stour Bay's practice was credible, but so far as the period prior to April 2015 is concerned, i.e., the period prior to the shortage of the pre-printed paper, their evidence was amply supported by the documentation in the bundles. So far as the Invoice History document referred to in paragraph 132 above is concerned:

i) 10 of the 11 invoices rendered to Provimi France had been disclosed by Provimi (each had a Provimi "received" stamp on the front). Every one of them had the Stour Bay T&Cs printed on the reverse. The one remaining invoice, number 5050, also had the Stour Bay T&Cs on the reverse;

ii) The two invoices issued to Provimi Spain (both of which were disclosed by Provimi Spain) had the Stour Bay T&Cs printed on the reverse. The evidence that Ms Sanz gave about this in the final sentence of paragraph 13 of her witness statement was wrong;

iii) Five of the six invoices issued to Provimi Poland had the Stour Bay T&Cs printed on the reverse. Mr Kala was wrong insofar as he suggested otherwise in paragraph 12 of his witness statement. The sixth, invoice 6385 dated 11 May 2015, did not, but it was issued during the period when Ms Morgan's workaround was in operation. Notably, four of the invoices disclosed by Provimi Poland were stamped on the back, i.e., over the Stour Bay T&Cs;

iv) 10 of the 11 invoices issued to Provimi UK (including those issued to it under its former name), including one produced by Provimi itself, had the Stour Bay T&Cs printed on the reverse. In relation to the one remaining invoice, there was no copy of the reverse of the invoice available.

141. I was handed during Ms Ansell, QC's oral opening a further schedule listing eight invoices for Vitamin D3 500 Feed Grade product supplied to non-Claimant Provimi companies.^[4] In five cases, there was no copy of the reverse of the invoice available. In the other three cases, where the bundle included a copy of the reverse, the Stour Bay T&Cs appeared.
142. It is obviously impossible to be certain in relation to those invoices where a copy of the reverse was not available, but the documents I have described support the evidence of Stour Bay's witnesses that, prior to the contracts of sale in 2015 under which the present dispute arises, Stour Bay's invariable or near invariable practice was to issue and send invoices to the Claimants (and to other Provimi companies) which bore the Stour Bay T&Cs on the reverse, and I so find.

(x) The 2015 contracts

143. On 5 January 2015 an email was sent to Mr Gibbons (copied to Ms Pickford) from the support@gep.com email address explaining that contract 2014003775 had been executed by Mr Amouroux that day.
144. The email was in the usual form set out above. Attached were:

i) Parent PARS Agreement, number 2014003775 for a quantity of 45,000kg at EUR 14.0/kg for delivery to Europe between 1 January and 30 June 2015. As with the other Parent PARS Agreements, it gave the Product Name as Vitamin D3 500 and included the PAC code 191910101; and

ii) The three documents referred to in the Parent PARS Agreement and said to be attached to it. These were the Provimi Gelatin Specification, the FBL Product Information Sheet, and a document entitled "rebate Fermenta.docx", which set out a quantity rebate agreed between Stour Bay and Provimi in December 2014 (a copy of this document appeared in the bundle adjacent to the 18 May amendment referred to below).

145. On 13 March 2015 the Parent PARS Agreement was amended in accordance with the baisse clause by the issuance of another Parent PARS Agreement with the same reference number stipulating a reduced price of EUR 13.50/kg. On 18 May 2015 a further revised document was issued, this time showing a price of EUR 11.00/kg. These two revisions referred to and were accompanied by the same three documents.
146. A number of Child PARS Agreements were subsequently issued in respect of allocations of product to different Provimi companies, each specifically referring to Parent PARS Agreement, number 2014003775. As with previous Child PARS Agreements, each referred to PAC Code 191910101 and each referred to and attached the Provimi Gelatin Specification and the FBL Product Information Sheet.
147. Focussing on the Child PARS Agreements issued in respect of the four Claimant companies, these were:

Date	Child PARS Agreement	Quantity	Delivery
Provimi France
14.1.15	2015000074	4,000kg	Jan – Mar 2015
24.3.15	2015000800	9,000kg	Apr – Jun 2015
19.6.15	2015001925	5,000kg	Jun – Jul 2015
Provimi Spain
12.1.15	2015000059	1,000kg	Jan - Mar 2015
19.6.15	2015001922	2,200kg	Jun - Aug 2015
Provimi Poland
24.3.15	2015000798	4,000kg	Apr - Jun 2015
19.6.15	2015001923	2,000kg	End July 2015
Provimi UK
20.1.15	2015000164	1,500kg	Jan – Mar 2015
24.3.15	2015000802	3,000kg	Apr – Jun 2015
19.6.15	2015001926	3,000kg	End July 2015

Quantities of product under Parent PARS Agreement, number 2014003775 were also allocated to Provimi Belgium, Ireland, Hungary and Netherlands.

148. In all these cases, purchase orders were issued by the relevant Provimi companies in respect of the quantities ordered or "called off" by them following the practice I described earlier, and invoices were rendered by Stour Bay to those companies once the product had been delivered. Details were set out in Schedule 1 to the Particulars of Claim, as amended in Schedule A to the Defence.
149. The invoices rendered by Stour Bay in respect of these sales were all issued during the period 8 April to 1 September 2015. As set out in a schedule handed up to me by Ms Ansell, QC during her oral opening entitled "Invoices forming part of the claim", most of these did not have the Stour Bay T&Cs printed on the reverse. There were two apparent exceptions:

i) Invoice 0228 issued to Provimi Spain – in fact a credit note – had the Stour Bay T&Cs on the reverse. This was issued on 11 June 2015, a Thursday, when Ms Morgan was not working, and where the invoice (which appears to have been manually created) would have been produced by Ms Pickford. As Ms Pickford explained, she had her own stash of the special pre-printed paper; and

ii) Invoice 6523 issued to Provimi UK. The front page of this invoice appeared in the bundle immediately followed by a page containing Stour Bay T&Cs. I was told by Ms Ansell, QC in her written closing submissions, however, that this was produced as two electronic documents, one containing the invoice and the other the terms and conditions.

150. As set out earlier, April to September 2015 fell within the period during which Ms Morgan and Ms Jackson explained Stour Bay had a problem with the supplier of its pre-printed paper, and where Ms Morgan was adopting her workaround, which included sending copies of the terms and conditions along with the invoices as separate documents.
151. The question in these circumstances is whether, although theses invoices did not have the Stour Bay T&Cs printed on the reverse, I am satisfied that, when sent to the Claimants, they would have been (and were) accompanied by a separate document containing the Stour Bay T&Cs. On the balance of probabilities I am so satisfied. I say that for two reasons.
152. First, and principally, Ms Morgan was an entirely credible witness, and I have no reason to doubt her evidence that her system "would always have been followed". It was not gainsaid by any evidence adduced by Provimi. Indeed, in relation to the practice of including the Stour Bay T&Cs on the reverse of invoices in the period prior to April 2015, the evidence of the two Provimi witnesses who sought to give evidence on that topic was wrong.
153. Secondly, Provimi's disclosure included copies of invoices with the Stour Bay T&Cs on the reverse, but it also included a number of pages where Stour Bay's T&Cs appeared as stand-alone documents. Stour Bay's solicitors, Clydes, raised this with Provimi's solicitors, Pinsent Masons, in a letter dated 18 January 2021, inviting Provimi to confirm that these were likely to have been attached to other invoices.
154. Pinsent Masons' reply on 11 February 2021, was that:

"The issue of whether Stour Bay's terms and conditions were attached to Stour Bay's invoices (with such invoices being post-contractual documents) will be addressed in Provimi's witness evidence."

None of Provimi's witnesses, however, provided an explanation for the existence of these stand-alone documents containing the Stour Bay T&Cs within the Claimants' disclosure. The presence of these documents is consistent with the workaround Ms Morgan describes having been followed.

(xi) The problems and complaints

155. Mr Fiedorow of Provimi Poland suggested in cross-examination that there may have been problems with the Vitamin D3 500 Feed Grade product supplied by Stour Bay prior to 2015, but I was shown no documentary evidence of any such problems or complaints. In the absence of any such documents, I infer that, if there were any problems or complaints prior to 2015, they cannot have been significant.^[5]
156. Provimi's witnesses suggested that the problem with its pre-mixture began to emerge in July 2015. Mr Gibbons said in his witness statement that he was told by Mr Piccolin at a trade fair at some time during 2015 that there was a potential issue that was being investigated, but it was September 2015 when the problem was first raised in correspondence with Stour Bay.
157. On 14 September 2015 Mr Piccolin sent an email to Mr Gibbons and Ms Jackson saying that they had some issues with a pre-mixture made by Provimi France containing vitamin D3 derived from FBL's Vitamin D3 500 Feed Grade product, asking Mr Gibbons to provide:

"… the manufacturing principle, the product composition, the composition of the gelatin used, the method of analysis recommended".

158. Mr Gibbons passed this request on to FBL. FBL's response was to ask for more information about the issues faced by Provimi so that its technical team could assist, but it attached to its email a number of the documents Mr Piccolin had asked for. FBL's email said, consistently with what Mr Piccolin had been told back in April 2010:

"Please note that our Vitamin D3 500 does not contain Gelatin"

and the composition details for its Vitamin D3 500 Feed Grade product that were attached did not identify gelatin as among the raw materials used.

159. Later on 15 September 2015 Mr Piccolin sent an email to both Stour Bay and FBL providing more details about the problem faced by Provimi, identifying Roseline Bobaz as the individual at Provimi France who was taking the lead:

"We have a big loss of D3 in a premix and it drives to a major trouble with our customer, as you can imagine. We are close of losing him.

The point is that after many investigations, we don't understand the root cause.

We have made the test with several batches of D3 Fermenta, confirming the bad results.

On the other side it is the only premix with whom we have this issue, among the one tested of course, as long as I understood.

Do you have any work, data about stability, interaction with other components of D3 500?"

160. Lucky Thakur, FBL's Site Quality Manager, responded to Mr Piccolin directly on the same day, attaching documents showing the specification and composition of FBL's Vitamin D3 500 Feed Grade product and (although missing in the bundle) a document setting out the manufacturing flow or process. In response to Mr Piccolin's question about stability, he explained:

"Regarding to product stability, our stability studies are on-going, and 18 months of valid stability data is available. Based upon this data we have assigned a shelf life of 24 months. We have also analysed our control samples after 24 months and found that all of them are fairly stable and meeting to the specification."

161. On 22 September 2015 and 30 September Mr Gibbons exchanged emails with Ms Bobaz. In response to a request from Mr Gibbons, Ms Bobaz provided details of the Provimi pre-mixture formulations used, emphasising that they were confidential. She said that Provimi had experienced bad results with each of them, even where the pre-mixtures had been produced only 10 days before they were analysed in a laboratory.
162. The emails in September 2015 described above concerned a problem experienced by Provimi France, but on 2 October 2015 Stour Bay received an email from Sara Allan of Provimi UK. Ms Allen explained:

"We are having problems detecting the vitamin D3 we purchase from yourselves once included in premixes and minerals, the raw material itself tests ok so I am thinking that there's something potentially interfering with the test. I have asked the labs for their opinion on the tests, but I am waiting to hear back from them. I'm wondering if you or the manufacturer of the vitamin D3 have ever come across this or would be able to help in anyway as to why this may happen.

Basically we have a customer complaining that they cannot detect it in the premix but they can in their finished feed and we have found the same … I think it may be one of those that could blow out of proportion if we do not resolve why it is happening sooner rather than later and would appreciate if you could help asap."

Stour Bay passed the enquiry on to FBL, who said that the issue was rather unusual and not one that they had come across before.

163. On 3 October 2015 Prashant Gudhate, Vice President Corporate Quality & Regulatory at FBL, contacted Mr Piccolin and explained the investigation that FBL had been carrying out itself, which he said he would share. Mr Gudhate said that he had never previously encountered the vitamin D3 content dropping by 95% within 10 days, and asked for information about Provimi's formulation and production processes.
164. On 19 October 2015 Mr Gudhate sent Mr Piccolin and Ms Bobaz the results of FBL's own investigation. This was a document entitled "Investigation Report of Customer Complaint". After reviewing various aspects of FBL's manufacturing process and summarising the results of re-analysis of control samples of the relevant batches, the document stated:

"Summary Report:

Based on complete review of mfg. process, input materials, testing method and reanalysis of the batches dispatched, we can conclude that there no potency drop of Vitamin D3 is seen in Vitamin D3 500 Feed. Same has also been confirmed by customer, as analysis of these batches at customer end found complying to specs.

Extended Investigation.

Even though as a standalone there is no quality issue with our product, however customer has noticed drop in Vitamin D3 content once 500 Feed batches are used in his premix. Therefore extended investigation is necessary to identify root cause of Vitamin D3 loss in customer's premix.

To support extended investigation following actions are planned,

1. Get understanding of Premix formulation of customer, ingredient composition and evaluate any possible impact on Vitamin D3 based on chemistry knowledge and other available information from literature.

2. Evaluate manufacturing process of Premix – to understand any potential of non homogeneity or degradation of Vitamin D3 during process operations. It could be done based on review of manufacturing operations as Customer's end. FBL representative shall visit manufacturing site for technical support. Subject to acceptance from Cargill for participation of FBL during investigation process at site.

3. Understanding of sampling process, assurance of sample represents the batch.

4. Understanding of analytical method used for analysis of Vitamin D3 in Premix.

5. Withdrawal of samples from multiple batches and perform analysis from two independent laboratories, one from EU and other from India. Number of samples per batch will be decided based on point no. 3."

165. Mr Piccolin provided Mr Gudhate with an update on 23 October 2015 explaining the claims and requests for compensation that Provimi had received. An email from Mr Gibbons to FBL on 3 December 2015 referred to a call with Provimi that morning in which he had been told that claims had been received from three customers in France, Spain and the UK amounting in total to around EUR 2.25 million.
166. On 8 and 9 December 2015 there was a meeting of representatives of Provimi, Stour Bay and FBL at the premises of Provimi France in Crevin. A combined Provimi/FBL PowerPoint presentation was prepared summarising the investigations so far carried out by each of them. Mr Fiedorow subsequently circulated within Provimi a summary of the discussion, which included the following:

"Conclusions and decisions:

Fermenta declared open approach to solve the issue and will for cooperation to investigate, and elucidate root cause (also if this confirms lack of stability of their product).

Final agreement concerning root cause accepted by both parties is to be developed as result of additional independent study, which details were agreed during the meeting. (Current position of Cargill is: the issue is caused by Fermenta product lack of stability in premix, current position of Fermenta/Stourbay is: the root cause of the issue is still to be confirmed).

Course of actions concerning agreed study that will elucidate root cause accepted by both parties was agreed.

There was also announced by Cargill team further actions concerning formal claim development and delivery to Stourbay (the claim will be further announced by Stourbay to their insurance provider).

Both actions: study and claim will be performed independently.

Further actions agreed and announced:

Root cause independent study:

Cargill and Fermenta will perform independent stability study using the same materials, formula, and agreed method

Cargill will deliver to Fermenta formula of premix (based on worst case notified in Cargill vit D3 stability tests), there will be also delivered chosen raw materials to produce small scale premix.

Using vit D3 material blocked in Crevin warehouse as well as new material delivered by Fermenta both companies will produce small scale premixes with listed vit D3 materials. Premixes will be analyzed in stability study by both parties independently using agreed reference method.

Cargill claim delivery

Cargill will deliver claim to direct supplier of vit D3 Stourbay. Stourbay will inform about the claim Stourbay insurance provider."

167. Separately, on 15 December 2015 Mr Fiedorow circulated amongst Provimi, Stour Bay and FBL a summary of the follow-up actions agreed during the meeting. This was consistent with the Root cause independent study section of the meeting note, and Mr Gudhate subsequently confirmed that it covered all the action items discussed at the meeting.
168. On 4 January 2016, as foreshadowed by the discussions at the Crevin meeting, Provimi France sent a letter to Stour Bay (marked "without prejudice" but included in the trial bundle without objection) formally notifying a claim on behalf of various Provimi entities relating to the (allegedly) defective vitamin D3 supplied by Stour Bay and giving an estimate of the customer claims.

(xii) The settlements

169. In the event, the first three Claimants, Provimi France, Provimi Spain and Provimi Poland, paid (or gave credit or supplied free product in) the following amounts to settle customer claims. Whilst Stour Bay did not accept that the settlements were reasonable, the amounts paid were documented and/or supported by Provimi's witnesses, and they did not appear to be in dispute:

i) Provimi France:

a) Nutrea: EUR 900,000 (against a total claim of EUR 1,232,442.35);

b) Fratelli Borello S.r.l.: EUR 60,000 plus EUR 127,000 (EUR 187,000 in total);

c) Fanin S.r.l.: EUR 350,000 plus EUR 250,000 (EUR 600,000 in total).

ii) Provimi Spain:

a) Huevos-Leon: EUR 300,000 plus product with free delivery to the sum of EUR 10,480.47 plus a credit note for used product to the value of EUR 47,255.18, plus the return of unused and unpaid for product to the value of EUR 8,386.47 (EUR 365,122.12 in total).

iii) Provimi Poland:

a) Syl-Drob: 115,000 zl (equivalent to EUR 26,000).

G. Terms and Conditions

170. As I indicated earlier, the issues between the parties fall into essentially three categories. The first category embraces two issues relating to the terms upon which the parties contracted, namely:

i) Did the contracts of sale incorporate and require compliance with the Provimi Gelatin Specification?

ii) Were the Stour Bay T&Cs incorporated into the contracts of sale by a course of dealing?

(a) An anterior issue?

171. It was common ground between the parties on the pleadings that, although a bulk order was initially placed by Provimi's central purchasing department, where that order was subsequently divided between different Provimi companies, the end result was that a number of separate contracts of sale were concluded between Stour Bay and each relevant Provimi company.
172. In paragraph 2(2) of its Reply, summarising its case as to the key documents by which it contended these separate contracts were created, Provimi said the following:

"(2) The key exchange of documents by which the particular contract was concluded would comprise the Claimants' Purchase Order (the offer) and the Defendant's email confirmation of the order or, where there was no such email confirmation, delivery of the Product (the acceptance of the offer)."

173. Mr Kulkarni, QC's written and oral submissions, however, put the position rather differently. In paragraph 1 of his written opening submissions, Mr Kulkarni, QC stated that:

"The present claim arises under a framework agreement in 2015".

In paragraph 43(c) of the same document, having referred to Parent PARS Agreement, number 2014003775, he continued:

"A binding contract between SB and Provimi was formed at this point: the parties had agreed volume, price, delivery method and the delivery period. All that remained was to allocate the agreed volume to particular Provimi businesses and to fix the particular deliveries to the individual businesses."

174. In his oral opening, when he took me to Parent PARS Agreement, number 2014003775 Mr Kulkarni, QC submitted that, "There's a clear statement of agreement here". Similarly, in his oral closing submissions (reflecting his written "closing roadmap"), he said:

"The parties considered there to be a binding contract upon the parent PARS agreement being issued and each of its children … I submit it's plain that everyone involved considered there to be a binding contract upon the parent PARS being accepted".

175. To the extent that, by these submissions, Mr Kulkarni, QC was seeking to rely upon Parent PARS Agreement, number 2014003775 as a stand-alone contract, as Ms Ansell, QC rightly observed in her own written opening submissions this was a fundamental change to the way in which the claim had been put. As she also rightly observed, a claim advanced on this basis would be hopeless:

i) In response to the rhetorical question posed by Ms Ansell, QC in opening, "Who was the contract contained in the Parent PARS Agreement between?" Mr Kulkarni, QC said in his closing submissions that it would have been between Provimi Netherlands, as the buying entity for Europe, and Stour Bay;

ii) But Provimi Netherlands – Provimi B.V. – was not a claimant in this action, and Provimi's case pleaded case is that it was the four Claimant companies that had contracted with Stour Bay and that had been faced with, and had settled, claims from customers and that had suffered loss.

176. In truth, Mr Kulkarni, QC's focus on the Parent PARS Agreement was really a forensic change of emphasis: an attempt, obviously with the issue of incorporation of the Stour Bay T&Cs in mind, to move the enquiry away from the dealings between Stour Bay and the individual Provimi companies – away from what Provimi's pleading identified as the key exchanges: the purchase order, the acknowledgment from Stour Bay, and the delivery of the product, following which an invoice, containing or accompanied by the Stour Bay T&Cs was issued – to the earlier Parent PARS Agreement concluded with Provimi's central purchasing department to whom invoices were never sent.
177. In saying this, I should not be understood to be suggesting that the Parent PARS Agreement did not have contractual effect; contrary to what Ms Ansell, QC suggested, it seems to me that it was plainly the intention of the parties that it should. However, the relevant contracts, and the contracts under which the present claims are made, are the contracts subsequently concluded by means of the exchanges identified between Stour Bay and the Claimant Provimi companies.^[6]
178. Mr Kulkarni, QC's change of emphasis also gave rise to a dispute as to the nature of the contracts of sale and how they were made, which was arguably of rather more importance.
179. In his oral opening submissions Mr Kulkarni, QC submitted that the contract – which, as indicated above, he suggested was Parent PARS Agreement, number 2014003775 was "a true blue written contract". It was, he submitted, therefore impermissible – and, indeed, a contractual heresy – for the court to look at the preceding exchanges for the purpose of ascertain the terms of or construing the contract.
180. This is not, however, a case where there was a written, formally executed contract document signed by both parties, and where the contract was a written contract in that sense:

i) It is obviously not the case when one considers the separate contracts concluded by the individual Provimi companies; Mr Kulkarni, QC's clients own pleaded case, set out in paragraph 172 above, makes that clear;

ii) But it would also not be true even if one were to focus solely on the initial "framework" agreement (Mr Kulkarni, QC's expression). As was common ground, the issuance of a Parent PARS Agreement would be preceded by agreement (either orally or by email) of basic terms between Mr Gibson and Mr Piccolin or Mr Amouroux. Mr Kulkarni, QC agreed in his oral closing submissions that, as a matter of English law, there would be a binding contract at that earlier stage. The Parent PARS Agreement and the email by which it was sent were, thus, intended to confirm an agreement already made.

181. Ms Ansell, QC's submission was that the relevant contracts were made partly orally, partly in (or at least partly evidenced by) writing – by the Parent PARS Agreement, the Child PARS Agreements, the purchase orders and any acknowledgement - and partly by conduct. The PARS Agreements, she said, did not constitute the full terms of an exclusively written contract. I agree with that submission.

(b) The Provimi Gelatin Specification

182. This takes me to the first of the two pleaded issues under this heading: whether the contracts of sale between the parties included, and required Stour Bay to ensure that the Vitamin D3 500 Feed Grade product complied with, the Provimi Gelatin Specification.
183. Mr Kulkarni, QC's case was straightforward. He submitted that the Provimi Gelatin Specification was clearly and expressly incorporated into each of the Parent and Child PARS Agreements by reference to the PAC Code, the specification itself accompanying the Parent and Child PARS Agreements.
184. Ms Ansell, QC's submission in response, consistent with her position as to the nature of the contracts and how they were formed, was that, in circumstances where the contracts of sale were not wholly written, the court was entitled to, and should, look at all the evidence from start to finish to see what the bargain was that was struck between the parties, evidence which she said included:

i) The 30 April 2010 email in which Mr Piccolin was told that the Vitamin D3 500 Feed Grade product manufactured by FBL was gelatin-free;

ii) The documentation to the same effect supplied by FBL as a part of the material evaluation and supplier approval process; and

iii) The fact, known to both Provimi and Stour Bay, that Provimi was only able to order a product that its Quality Assurance department had previously approved; and the fact that the only FBL Vitamin D3 500 Feed Grade product that Provimi ever had approved did not contain a gelatin (or any) coating.

185. Ms Ansell, QC relied for her "start-to-finish" proposition on the decision of the Court of Appeal in J. Evans & Son (Portsmouth) Ltd v Andrea Merzario Ltd [1976] 1 WLR 1078.
186. Evans v Merzario concerned a contract between an English importer and a freight forwarder in relation to the carriage of a machine. The forwarder's standard terms gave it complete freedom as to how the machine was transported, but the parties had agreed orally that, if the machine was carried in a container, the container would not be shipped on deck. A mistake was made, the container with the machine was shipped on deck, and it was lost.
187. The case was dealt with at first instance on the basis that the relevant question was whether there was a binding collateral warranty, a warranty collateral to the contract between the parties which was on the freight forwarder's standard terms, that any container would not be carried on deck. Kerr J decided that there was no such collateral warranty. Lord Denning MR appears to have dealt with the case on the same basis, but he disagreed.
188. The two other judges in the Court of Appeal, Roskill and Geoffrey Lane LJJ agreed with Lord Denning MR that the appeal should be allowed, but (and although Roskill LJ stated that he agreed with the reasons given by the Master of the Rolls) their judgments indicate that they approached the matter on a somewhat different basis. At 1083C-H, in a passage the final sentence of which was relied upon by Ms Ansell, QC, Roskill LJ said this:

"The matter was apparently argued before the learned Judge on behalf of the plaintiffs on the basis that the defendants' promise (if any) was what the lawyers sometimes call a collateral oral warranty. That phrase is normally only applicable where the original promise was external to the main contract, that main contract being a contract in writing, so that usually parol evidence cannot be given to contradict the terms of the written contract. … But that doctrine, as it seems to me, has little or no application where one is not concerned with a contract in writing … but with a contract which, as I think, was partly oral, partly in writing and partly by conduct. In such a case the Court does not require to have recourse to lawyers' devices such as a collateral oral warranty in order to seek to adduce evidence which would not otherwise be admissible. The Court is entitled to look at and should look at all the evidence from start to finish in order to see what the bargain was that was struck between the parties."

189. The judgment of Geoffrey Lane LJ at 1084G-H was to similar effect:

"I agree, for the reasons already expressed, that the effect of the conversation between Mr. Spano of the defendants and Mr. Leonard of the plaintiffs in the autumn of 1967 was to produce a binding obligation on the defendants to ensure that the plaintiffs' machinery in containers would be carried under deck between Rotterdam and Tilbury. This was not a collateral contract in the sense of an oral agreement varying the terms of a written contract. It was a new express term which was to be included thereafter in the contracts between the plaintiffs and the defendants for the carriage of machinery from Italy."

190. Ms Ansell, QC also relied upon a passage in the judgment of Donaldson J in S.I.A.T. di del Ferro v Tradax Overseas SA [1978] 2 Lloyd's Rep 470 (affirmed at [1980] 1 Lloyd's Rep 53), a case concerning incorporation of standard terms and conditions, at page 490 where Donaldson J said:

"… a contract is not made in a vacuum, but against a background of present and past facts and future expectations and that its terms, and indeed the consensus itself, are to be gathered not only from expressed words but also from conduct viewed against that background".

191. On the basis of this "start-to-finish" analysis, Ms Ansell, QC submitted that the express terms of the contracts of sale were that:

i) Stour Bay would supply the Vitamin D3 500 Feed Grade product sourced from FBL, as supplied in the samples and trials and approved by Provimi, and as identified in the FBL Product Information Sheet;

ii) Stour Bay would supply FBL's Vitamin D3 Feed Grade product at the prices negotiated by Mr Gibbons with Mr Piccolin from time to time in respect of overall quantities; and

iii) Stour Bay would supply batches of that overall quantity to the individual Provimi companies in the quantities, and on the dates and to the addresses set out in the purchase orders that those individual companies placed.

192. The Provimi Gelatin Specification, she submitted, was not part of the express terms of the contract of sale. It was not provided to Stour Bay until after the first order had been agreed; it was inconsistent with the 30 April 2010 email and the questionnaires, and with what was or should have been known by Provimi, which made clear the absence of gelatin coating; and it was inconsistent with the FBL Product Information Sheet.
193. Importantly, Ms Ansell, QC submitted, if the Provimi Gelatin Specification was incorporated into the contracts of sale it would lead to the following absurdity: either

i) The parties contracted on the basis of the Provimi Gelatin Specification knowing from the outset that the FBL Vitamin D3 500 Feed Grade product did not comply with it; or,

ii) In circumstances where both parties understood that Provimi could only purchase a product that had been pre-approved, compliance with the Provimi Gelatin Specification required Stour Bay to supply a product different to the only product that Provimi had, in fact, approved, which did not contain a gelatin (or any) coating.

194. For completeness, Ms Ansell, QC submitted that, even if I were to decide (contrary to her primary case) that the Parent PARS Agreements were complete bi-lateral contracts, on ordinary principles of construction I should conclude that the contracts were for the sale and purchase of the FBL Vitamin D3 500 Feed Grade product, and not for the sale and purchase of (what would be a different) product that complied with the Provimi Gelatin Specification.
195. I do not consider this issue to be an easy one. I am conscious, in particular, of the frequency and consistency with which the Provimi Gelatin Specification was referred to in, and sent to Stour Bay as an attachment to, the Parent PARS and Child PARS Agreements, a point that Mr Kulkarni, QC justifiably emphasised. By 2015, when the relevant contracts of sale were made, the specification had been referred and sent to Stour Bay a number of times.
196. On balance, however, I prefer Ms Ansell, QC's submissions. I consider that the Provimi Gelatin Specification was not part of the agreed contract terms, and that compliance with it by Stour Bay – in particular, by supplying a Vitamin D3 500 Feed Grade product with a coating of non-ruminant gelatin - was not required. I reach that conclusion for the following principal reasons.
197. First, as I indicated earlier, the contracts of sale made between Stour Bay and the Claimants in 2015, and those made with the same companies and with other Provimi companies in previous years, were contracts that were made partly orally, partly in writing, and partly by conduct. I agree, in these circumstances, that in determining what the terms of the contracts were I can and should look at the whole of the parties' dealings.
198. Secondly, I regard the circumstances in which the Provimi Gelatin Specification first came to be provided to Stour Bay as significant:

i) As I explained earlier, the basic terms of the first order placed by Provimi were agreed in the email exchanges between Mr Gibbons and Mr Piccolin on 15 and 18 February 2013, those emails concluding what Mr Kulkarni, QC himself accepted would be a binding contract;

ii) At no stage, either in those emails or at any prior time, was Mr Gibbons (or Stour Bay) provided by Mr Piccolin with the Provimi Gelatin Specification or told by Mr Piccolin that what was being ordered by Provimi was a product that complied with this specification;

iii) The Parent PARS Agreement sent to Stour Bay by email to which the Provimi Gelatin Specification was attached did not purport to be an offer for Mr Gibbons to accept or reject - there was no evidence that it was regarded in that way – but a confirmation of a contract already agreed; a contract, as I have just explained, that had been made without any reference or regard to that specification.

199. The Parent PARS Agreement did, of course, refer to and attach the Provimi Gelatin Specification. Mr Gibbons said that in his witness statement that he assumed that the document was a template or that it had been issued in error. He said that he ignored it because, so far as he was concerned, Provimi was purchasing the FBL Vitamin D3 500 Feed Grade product that it had recently approved.
200. Mr Kulkarni, QC criticised this evidence, saying that, if it was true, it exhibited a reckless approach to contracting. He made the point, by reference to a passage in Ms Jackson's witness statement, that what would usually be expected of a manufacturer such as FBL (and I assume he would say the same of a supplier such as Stour Bay) would be that, if it was sent a specification, it would speak up if it had any issue with producing goods to that specification.
201. Ultimately, though, as Mr Kulkarni, QC said himself, whether Mr Gibbons might be criticised for not querying the Provimi Gelatin Specification, what is ultimately relevant is whether its application had been agreed and whether Stour Bay was bound to comply with its terms. It is relevant in that regard that the Parent PARS Agreement not only referred to and attached the Provimi Gelatin Specification, but – reflecting that the product being purchased was the FBL product – it also referred to and attached the FBL Product Information Sheet, which was in at least one respect inconsistent.
202. Thirdly, whilst the nature of the Provimi Gelatin Specification is such that it is no doubt capable of being used by Provimi as a contractual specification agreed with a third-party supplier, the passage in Mr Piccolin's cross-examination that I set out in paragraph 62 above suggests that the PAC number and specification had an administrative, internal function. So far as this is concerned:

i) No communications were disclosed between the Claimants and Provimi's central purchasing arm showing the terms in which the Claimants indicated their requirements for Vitamin D3 Feed Grade product that led to the orders placed with Stour Bay. It may be that these were emails that were deleted pursuant to the Cargill document retention policy described earlier;

ii) But internal, administrative reasons may provide an explanation as to why, not having been mentioned at all by Provimi during its negotiations with Stour Bay, the Provimi Gelatin Specification then came to be attached to the Parent and Child PARS Agreements, alongside the FBL Product Information Sheet as was appropriate given that it was the FBL product that was being ordered.

203. Fourthly, there is Provimi's requirement for material and supplier approval. Whilst these requirements were Provimi's own, they were explained by Mr Piccolin to Mr Gibbons on a number of occasions prior to the contracts of sale being concluded in which it was made clear that approval of the product was a pre-condition to any order being placed. It is, in my judgment, an important background fact, known to both parties at the time they contracted, that Provimi could only order a product that it had first evaluated and approved.
204. The evidence in that regard is clear: the only FBL Vitamin D3 500 Feed Grade product that had been submitted for approval to Provimi was a product that did not contain a gelatin (or any) coating. Mr Piccolin had been told that the produce was gelatin-free in April 2010, and even if that might be regarded as distant in time from the orders placed, the information provided to Provimi's Quality Assurance department in 2011 and 2012 in relation to the approval process indicated the same.
205. As Mr van der Eijk accepted – see paragraph 101 above – it was obvious from the information supplied to the Quality Assurance department that the product did not have a gelatin coating. I did not hear any evidence from Mr Brouwer or Mr Shekhar, but in light of Mr van der Eijk's answer, and in light of the common ground between the experts, as I said earlier I conclude that this is a fact that they must have appreciated at the time.
206. Mr Kulkarni, QC's submissions in response to the suggested absurdity referred to in paragraph 193 above, were, with respect to him, unpersuasive:

i) He suggested that suppliers could be expected continually to develop products, so that a product range in April 2010 (the date of the email to Mr Piccolin) would not necessarily be the same as the product range a year or two later. That may be true, but it is no answer to the fact that the only product submitted to Provimi for approval, and the only product actually approved by it, was a product without a gelatin (or any) coating;

ii) He submitted that the Provimi approval process was for Provimi's benefit, and that if Provimi had already approved a supplier and one of its products there was no reason why it could not subsequently choose to order a different product from that same supplier. This submission, however, is at odds with the evidence about Provimi's approval system that I summarised earlier: see, in particular, paragraph 64 above. It does not appear to reflect what either party understood at the time, namely that approval of the product to be supplied was essential.

207. Mr Kulkarni, QC's submission also sits unhappily with the following passage in Mr Piccolin's cross-examination in which Mr Piccolin confirmed that the only product he was able to order, and the product which he asked Stour Bay to supply, was the FBL product that had been approved:

"Q. So the only vitamin D3 powder product that Stour Bay was authorised or could supply to Provimi was this PFB Vitamin D3 500 Feed Grade which had been approved?

A. Yes.

Q. And so all the negotiations thereafter concerned the approved FBL vitamin D3 Feed Grade product?

A. When we were talking about D3 500, yes.

Q. Yes. That was what you were offering to purchase because there was no other vitamin D3 500 powder which Stour Bay was permitted to provide or supply to Provimi?

A. Yes.

Q. And that is what you asked FBL – sorry, that's what you asked Stour Bay to quote for, that product, when you were talking about vitamin D3 500?

A. I suppose yes.

…

Q. So, the only product that Stour Bay could supply in accordance with this document [the Parent PARS Agreement for the first order] would be the product produced by FBL?

A. Yes, according to the sample sent by Stour Bay for our approval procedure."

208. In my judgment, against this background, a conclusion that the Provimi Gelatin Specification was incorporated into the contracts of sale, the result of which would be that the parties had contracted to buy and sell a product different to that which had been sent to Provimi for approval, and which Provimi had approved, would be at odds with any objective analysis of the totality of the dealings between the parties and cannot be justified.
209. Fifthly, I do not consider that the fact that the Provimi Gelatin Specification was sent as an attachment to each Parent and Child PARS Agreement can sensibly be regarded as changing the status that the document had at the time it was attached to the first. Mr Kulkarni, QC rightly points out that the relevant contracts were those made in 2015, but in my judgment the Provimi Gelatin Specification did not form part of the terms of those contracts or any of the preceding contracts between the parties for the purchase of Vitamin D3 500 Feed Grade product.

(c) The Stour Bay T&Cs

210. The second issue that arises under this heading concerns the incorporation of the Stour Bay T&Cs. Stour Bay's case was that they were incorporated into the contracts of sale made in 2015 as the result of a course of dealing.
211. There was some debate between the parties as to exactly what was required to constitute a sufficient course of dealing. The parties referred me to a large number of authorities, some said to demonstrate the principles but others relied upon simply as illustrations of circumstances which had or had not been held to be sufficient for terms to be incorporated.
212. With respect to the parties, I do not propose to lengthen this judgment by dealing with each and every one of these authorities, many of which are well-known. The overarching test is that set out by Lord Pearce in Henry Kendall & Sons v William Lillico (the Hardwick Game Farm case) [1969] 2 AC 31 at 113D who, referring to the previous dealings between the parties, said this:

"The court's task is to decide what each party to an alleged contract would reasonably conclude from the utterances, writings or conduct of the other."

213. I was referred by both parties to the decision of Christopher Clarke J in Balmoral Group Ltd v Borealis [UK] Ltd [2006] EWHC 1900 (Comm), [2006] 2 Lloyd's Rep 629. That was a case, like this one, where reliance was placed on the inclusion by one party of its standards terms and conditions on the reverse of invoices, and to that extent the case was useful.
214. Christopher Clarke J commenced his discussion at [348] with a salutary warning as to the relevance of prior authority:

"Whether or not one party's standard terms are incorporated depends on whether that which each party says and does is such as to lead a reasonable person in their position to believe that those terms were to govern their legal relations. The court has to determine what each party was reasonably entitled to conclude from the acts and words of the other: McCutcheon v David Macbrayne Ltd [1964] 1 WLR 125, 128; Kendall & Sons v Lillico & Sons [1969] 2 AC 31; Hollier v Rambler Motors (AMC Ltd) [1972] 2 QB 71. The question is one of fact to which prior authority may form an uncertain guide."

215. Thereafter, at [356], having reviewed a number of authorities (including many of the cases referred to by Mr Kulkarni, QC and Ms Ansell, QC in their submissions), Christopher Clarke J summarised the position in the following way:

"These cases show two things. Firstly, at any rate where parties have dealt with each other more than once or twice, it may not be critical to the incorporation of standard terms that those terms be set out in a contractual document, ie one that itself constitutes an offer or its acceptance, or even in a purported record of the contract, nor that the document containing the terms relied on has preceded the making of every contract. Secondly, the sequence of events is important. An invoice following a concluded contract effected by a clear offer on standard terms which are accepted, even if only by delivery, will or may be too late. But, if there has been no reference to rival terms, the appearance of terms on the back of every invoice and the acceptance of delivery of goods without objection may indicate acceptance of the terms."

216. The two points made by Christopher Clarke J in this passage are of some relevance to this case.
217. So far as the first is concerned, Mr Kulkarni, QC said that Stour Bay's invoices were sent after the product had been delivered and were post-contractual, by which I understood him to mean that they were generated in the performance of the contract, triggering the relevant Provimi company's obligation to pay, not in the contracts' creation. That is true; but as Christopher Clarke J explained, depending on the nature and frequency of the parties' dealings, this may not be a critical factor.
218. As for the second point, as Christopher Clarke J said, the context in which invoices and their accompanying terms and conditions appear is important. He distinguished between a situation where an invoice containing terms and conditions followed a contract "effected by a clear offer on standard terms which are accepted" from a situation where there were no such rival terms.
219. Mr Kulkarni, QC relied in this context upon the message in the standard form emails that accompanied the PARS Agreements:

"Attached you can find our Agreement containing terms and conditions for the coming deliveries of your products to our affiliates. Individual Purchase Orders and delivery details will be issued by local Cargill entities, in accordance with this agreement";

He said – correctly – that Stour Bay never queried the application of those terms and conditions.

220. This point, however, is, in my judgment, of limited weight. The message in the standard email said no more than that the Agreement attached contained "terms and conditions" – which it undoubtedly did – not that it contained "the" or "all the" terms and conditions. As a matter of fact, what the PARS Agreements contained were no more than the basic terms agreed in the prior email or telephone exchanges: quantities, price, delivery period, and payment terms. This is not a case, like many, where there was anything approaching a complete set of rival terms and conditions.
221. Dealing with the facts of the particular case before him, at [358]-[359] Christopher Clarke J said the following (which includes a passage on which Ms Ansell, QC placed reliance):

"358. … By putting its terms on the back of their invoices Borealis indicated to Balmoral that, so far as Borealis were concerned, they intended the contract to be on those terms: see clause 2 cited at para 30 above. Those invoices were seen and initialled by Mr Joyce, the Managing Director of Balmoral's rotomoulding division. He realised that there were terms on the back of the invoices. He reviewed them but did not study them. Balmoral thereafter purchased material at the quoted prices with knowledge of Borealis' conditions without ever suggesting that they were not applicable. Borealis did not know that each invoice in respect of borecene was initialled by Mr Joyce before payment. But they did know that they were received and paid without demur. A reasonable person in their position would be entitled to assume, as was the fact, that someone at Balmoral had considered whether or not to pay the invoice and had seen the conditions on the back.

359. In those circumstances Borealis was, in my judgment, reasonably entitled to assume that Balmoral accepted the applicability of the conditions subject to which the price had been quoted. If Balmoral wanted to pay the price quoted without accepting the terms, it was incumbent on it to say so."

222. I was also referred to the decision of Edwards-Stuart J in Transformers & Rectifiers Ltd v Needs Ltd [2015] EWHC 269 (TCC), [2015] BLR 336, in particular in relation to the number and consistency of prior transactions that were necessary to constitute a sufficient course of dealing. At [42] Edwards-Stuart J summarised the principles he derived from prior authorities in the following way:

"42. From my rather brief review of some of the relevant authorities, I consider that in cases of this sort the following principles apply:

i) Where A makes an offer on its conditions and B accepts that offer on its conditions and, without more, performance follows, the correct analysis, assuming that each party's conditions have been reasonably drawn to the attention of the other, is that there is a contract on B's conditions: see Tekdata.

ii) Where there is reliance on a previous course of dealing it does not have to be extensive. Three or four occasions over a relatively short period may suffice: see Balmoral at [356] and Capes (Hatherden).

iii) The course of dealing by the party contending that its terms and conditions are incorporated has to be consistent and unequivocal: see Sterling Hydraulics.

iv) Where trade or industry standard terms exist for the type of transaction in question, it will usually be easier for a party contending for those conditions to persuade the court that they should be incorporated, provided that reasonable notice of the application of the terms has been given: see Circle Freight.

v) A party's standard terms and conditions will not be incorporated unless that party has given the other party reasonable notice of those terms and conditions: see Circle Freight.

vi) It is not always necessary for a party's terms and conditions to be included or referred to in the documents forming the contract; it may be sufficient if they are clearly contained in or referred to in invoices sent subsequently: see Balmoral at [352], [356].

vii) By contrast, an invoice following a concluded contract effected by a clear offer on standard terms which are accepted, even if only by delivery, will or may be too late: see Balmoral at [356]."

223. The same judgment at [44] and [45] makes clear that, whilst a buyer who wishes to incorporate his own terms and conditions must give the seller reasonable notice of those terms and conditions, this can be done be setting out the terms and conditions on the reverse of an invoice, so long, that is, as the seller is sent both the face and the reverse of the document.
224. Not included in Edwards-Stuart J's summary of principles, but clearly established by S.I.A.T. di del Ferro and by the decision of Colin Edelman, QC (sitting as a Deputy High Court Judge) in Lisnave Estaleiros Navais SA v Chemikalien Seetransport [2013] EWHC 338 (Comm), [2013] 2 Lloyd's Rep 203 at [28], when the contracting parties form part of a wider group executing similar transactions a course of dealing can be established by reference to transactions with the whole group.
225. So far as the facts are concerned, Ms Ansell, QC relied principally upon the number and consistency of prior transactions between Stour Bay and Provimi – the history of dealings that I summarised earlier in this judgment – but she also relied upon the fact, as she submitted, that terms akin to the Stour Bay T&Cs were usual in the animal feed industry, and indeed were used by Provimi companies themselves when selling product to third parties.
226. There is, of course, a well-known principle, for which the decision in British Crane Hire Corporation Ltd v Ipswich Plant Hire Ltd [1972] 1 WB 303 is usually cited as authority, whereby terms that are usual in the relevant industry can be incorporated into a contract even absent a course of dealing. As [42 iv)] in Transformers shows, however, the existence of industry practice can also support a course of dealing case. The reality, as Donaldson J explained in S.I.A.T. di del Ferro, is that incorporation of terms as a result of a course of dealing or by application of the British Crane doctrine are not two separate principles but simply facets of the wider concept that a contract is not made in a vacuum.
227. Ms Ansell, QC referred me to the standard terms used by Provimi France and by Provimi UK when they sold product, which were printed on the back of their invoices, as well as a number of sets of terms used by other industry participants such as Vitafor, NHU, Direct Foods and Lehvoss. These were not identical to the Stour Bay T&Cs, but they bore a number of similarities including the presence of limitation provisions. Mr Amouroux was asked about Provimi France's terms in cross-examination:

"Q. Now, I hope you see on your screen – and we'll get that blown up – standard terms and conditions which Provimi France use when they're selling a product. Are you familiar with –

A. Yes. Familiar. I know that – I know that they exist, but I won't say I'm extremely familiar with them.

Q. So when Provimi France is selling product, it sells on the basis of those terms and conditions which it prints on the back of its invoices.

A. Yes.

Q. And it is common in the industry, isn't it, for sellers to do as Provimi France do, which is to trade on standard terms and conditions of the type we can see on the screen, which they print on the back of their invoices?

A. Yes.

Q. And so what I was putting to you was that it was therefore – you must have known that it was standard in the industry for the supplier to have such terms and conditions and you must have known that Stour Bay had such terms and conditions.

A. If it was shown on the back of the invoice, I would, yes. These conditions are generally printed on the back of an invoice, but I did not know all the conditions from Stour Bay because generally I was not in contact with invoices per se."

228. I have touched on two of the points made by Mr Kulkarni, QC already: his submission that the invoices were post-contractual, and his reliance on the terms of the emails accompanying the PARS Agreements. Mr Kulkarni, QC also stressed that, although the Stour Bay T&Cs may have appeared on the reverse of invoices, there was no reference to them on the front, although as the relevant question is whether, on the facts of the particular case, reasonable notice of the terms and conditions has been given I do not regard that as in any way fatal.
229. Generally, Mr Kulkarni, QC submitted that Stour Bay's practice of sending its terms and conditions was inconsistent, and that there was no sufficient or consistent course of dealing to justify a conclusion that the parties must have intended to incorporate the Stour Bay T&Cs. He made the point that of the 17 invoices submitted for product supplied under Parent PARS Agreement, number 2014003775, only one (or two, if one includes the credit note) had the Stour Bay T&Cs on the reverse.
230. Mr Kulkarni, QC also took me to a passage in Lewison, The Interpretation of Contracts (7th ed), paragraph 3.108 where the author states:

"The question of whether terms have been incorporated by reference to a previous course of dealing is essentially a question of the implication of terms. Thus the usual tests for the implication of terms will need to be satisfied."

On the basis of the statement that the "usual" tests for the implication of term apply, Mr Kulkarni, QC referred me to the decision of the Supreme Court in Marks and Spencer plc v BNP Paribas Securities Services Trust Company (Jersey) [2015] UKSC 72, [2016] AC 742 and to Lord Neuberger PSC's remark at [21] that, whilst business necessity and obviousness were alternative criteria for implication:

"I suspect that in practice it would be a rare case where only one of the two requirements would be satisfied."

231. On the back of this remark, Mr Kulkarni, QC submitted in his written opening submissions that the requirement of necessity was not met because the contracts of sale between the parties were both commercially and practically coherent without the Stour Bay T&Cs. So, he suggested, there was no basis for implying the Stour Bay T&Cs into them.
232. Mr Kulkarni, QC, however, stepped away from this submission in his closing submissions, and rightly so. Whilst necessity may be the appropriate touchstone in some situations, most obviously where terms are sought to be implied into what is, on its face, a complete written agreement, it has not been treated as the relevant criterion in the course of dealing cases, and indeed in many cases where terms have been implied on the basis of a course of dealing (commonly one party's standard terms) it would be unlikely to have been satisfied.
233. The more appropriate criterion in course of dealing cases, in my judgment, is obviousness, by which I mean that, looking at the matter objectively, it must be obvious from the parties' dealings, that they intended the relevant terms and conditions to apply, the important point being that a conclusion that the parties might (but equally might not) have intended the terms and conditions to apply is not sufficient. This, it seems to me, is consistent with the authorities:

i) In S.I.A.T. di del Ferro at page 490, Donaldson J said that the test was:

"Would the parties have agreed that a particular term formed part of the contract if they were reasonable men looking at the matter objectively in the knowledge that no adverse consequences could flow from the answer. I add this latter qualification because even the most reasonable commercial man will become a little less than objective when told that the answer could cost him U.S. $3 million. The term will only be contractual if the parties' answer would have been a definite 'Yes'. 'Possibly' will not do";

ii) In Hamad M. Aldrees & Partners v Rotex Europe Limited [2019] EWHC 574 (TCC), having referred to Donaldson J's judgment in S.I.A.T. di del Ferro, at [180] Edwards-Stuart J said:

"Applying the test suggested by Donaldson J in the SIAT di del Ferro v Tradax Overseas case to the circumstances set out in the previous paragraph, I would expect the reasonable businessman to conclude that Rotex's terms and conditions were possibly incorporated, but not that they were definitely incorporated. But, as Donaldson J pointed out, 'possibly' is not good enough;"

iii) In Lisnave at [32] Colin Edelman, QC concluded his review of the authorities by saying:

"The question is not therefore whether it would have been reasonable for the parties to have agreed to incorporate into the Fleet Agreement the arbitration clause in the General Conditions. Rather, the question is whether the parties to the Fleet Agreement must have intended to include the arbitration clause in the General Conditions."

234. As all the authorities I have discussed make clear, whether or not terms are incorporated by a prior course of dealing is ultimately a question of fact and degree, which depends, amongst other things, on the number of previous contracts, how recent they are, the similarity of subject matter, and the manner in which they were concluded. A prior course of dealing does not require complete identity of companies.
235. In the present case, considering the evidence as a whole, I am satisfied that there was a sufficient course of dealing between the parties, that reasonable notice was given to Provimi (including the Claimant companies) of the Stour Bay T&Cs, and that the Stour Bay T&Cs were incorporated into the contracts of sale between the parties. My answer to the question posed by Colin Edelman, QC in Lisnave – whether, looking at the matter objectively, the parties must have intended that the contracts would be made on the Stour Bay T&Cs – is "yes".
236. My reasons are briefly these.
237. First, this is a case where there was a significant course of dealing between Stour Bay and Provimi, extending over around five years (between 2009 and 2015) prior to the conclusion of the relevant contracts of sale. The invoice history that I referred to in paragraph 132 above suggests that in the case of the Claimant companies alone there were in the order of 28 prior transactions over this period.
238. Secondly, and as I have found, Stour Bay's practice throughout this period was consistent: its terms and conditions were invariably or almost invariably printed on the reverse of its invoices, which appear to have been received and paid by the relevant Provimi companies without demur. The test of whether reasonable notice was given is an objective one, but Provimi Poland at least must have noticed the terms because it applied its stamp over them.
239. Thirdly, clause 2.2 of the Stour Bay T&Cs, the terms of which I set out in paragraph 41 above, made clear that Stour Bay intended to contract on its terms and conditions to the exclusion of all others. The position is, in that respect, similar to the position described by Christopher Clarke J in his judgment in Balmoral at [358] which I set out in paragraph 221 above.
240. Fourthly, this is not a case where, in reality, there was a rival set of terms and conditions. I have not overlooked the standard form email accompanying the PARS Agreement, but I do not regard this, or the documents attached to it, as inconsistent with the application of the Stour Bay T&Cs. The email, of course, specifically said that purchase orders would be issued by individual Provimi companies, and it contemplated that, in relation to the time allowed for payment at least, locally agreed terms might take precedence.
241. Fifthly, my conclusion that the Stour Bay T&Cs were intended to apply is bolstered by the fact that terms akin to those used by Stour Bay appear routinely to be used by other companies in the animal feed industry, including by at least two of the four Claimant Provimi companies (the evidence in relation to Provimi Spain and Provimi Poland was unclear).

H. Quality

242. My conclusions in relation to the incorporation of the Provimi Gelatin Specification and the Stour Bay T&Cs are determinative of Provimi's claim for damages in these proceedings, which must fail.
243. In deference to the substantial amount of evidence and submissions that I heard, and in case this matter goes further, I will, however, deal in this section and the next with the other issues – with the issues of breach, causation and loss.

(i) Compliance with the Provimi Gelatin Specification

244. There is one issue that I can dispose of straightaway, because the position is common ground.
245. The first of the two alleged breaches of contract concerns the Provimi Gelatin Specification and its requirement that the Vitamin D3 500 Feed Grade product sold by Stour Bay to Provimi should have a coating of non-ruminant gelatin.
246. As I have found, the Provimi Gelatin Specification was not incorporated into the contracts of sale, and Stour Bay was not required to comply with its terms. If that is wrong, however, then Stour Bay was plainly in breach: the product supplied did not have a coating of non-ruminant gelatin (or any coating).

(ii) Compliance with the SGA implied terms

247. As Mr Kulkarni, QC accepted, if the Stour Bay T&Cs were incorporated into the contracts of sale, as I have found that they were, then – and as is permitted by section 55 of the SGA - the statutory implied terms under section 14 of the SGA were excluded. The question of breach of those terms, accordingly, does not arise.
248. In case I am wrong about the incorporation of the Stour Bay T&Cs, however, I need to deal with Provimi's case that the quality and fitness for purpose of the Vitamin D3 500 Feed Grade product was such that Stour Bay breached sections 14(2) and (3) of the SGA. Those sections, insofar as relevant, provide as follows:

"(2) Where the seller sells goods in the course of a business, there is an implied term that the goods supplied under the contract are of satisfactory quality.

(2A) For the purposes of this Act, goods are of satisfactory quality if they meet the standard that a reasonable person would regard as satisfactory, taking account of any description of the goods, the price (if relevant) and all the other relevant circumstances.

(2B) For the purposes of this Act, the quality of goods includes their state and condition and the following (among others) are in appropriate cases aspects of the quality of goods—

(a) fitness for all the purposes for which goods of the kind in question are commonly supplied,

…

(e) durability.

(2C) The term implied by subsection (2) above does not extend to any matter making the quality of goods unsatisfactory—

(a) which is specifically drawn to the buyer's attention before the contract is made,

(b) where the buyer examines the goods before the contract is made, which that examination ought to reveal, or

(c) in the case of a contract for sale by sample, which would have been apparent on a reasonable examination of the sample.

(3) Where the seller sells goods in the course of a business and the buyer, expressly or by implication, makes known—

(a) to the seller, or

…

any particular purpose for which the goods are being bought, there is an implied term that the goods supplied under the contract are reasonably fit for that purpose, whether or not that is a purpose for which such goods are commonly supplied, except where the circumstances show that the buyer does not rely, or that it is unreasonable for him to rely, on the skill or judgment of the seller or credit-broker.

…

(6) As regards England and Wales and Northern Ireland, the terms implied by subsections (2) and (3) above are conditions."

(a) Provimi's case

249. I summarised Provimi's pleaded case as to the breach of these implied terms in paragraph 12 iii) above and Stour Bay's response to that case in paragraph 18. As is often the case, the allegations made under sections 14(2) and 14(3) of the SGA substantially overlapped.
250. In essence, Provimi asserts that, as a result of the manufacturing process and/or the absence of a coating, the Vitamin D3 500 Feed Grade product supplied by Stour Bay was insufficiently stable, durable, and/or robust with the result that the vitamin D3 content was liable to deteriorate or degrade to an unacceptable level as a result of ordinary processes, including mixing with commonly used pre-mixture ingredients.
251. In those circumstances, Provimi says:

i) The Vitamin D3 500 Feed Grade product was not of satisfactory quality for the purposes of section 14(2) of the SGA, the incorporation of the product into an animal (and a poultry) pre-mixture, Provimi submits, being one of the purposes for which a product of that kind is commonly supplied (see section 14(2B)(a)); and

ii) For the same reason, the Vitamin D3 500 Feed Grade product was not reasonable fit for the particular purpose for which it was purchased, a purpose implicitly made known to Stour Bay, namely for incorporation into an animal (and a poultry) pre-mixture.

252. So far as the case under section 14(3) is concerned, there was some debate as to the "particular purpose" that Provimi said was expressly or implicitly made known to Stour Bay.
253. Provimi's pleaded case (see paragraph 12(3) of its Particulars of Claim) was that the product had to be:

"… reasonably fit for the purpose of adding it to the Claimant's Premixture and/or industry standard pre-mixture".

Mr Kulkarni, QC confirmed in his oral submissions, however, that it was not his case that Stour Bay knew that the Vitamin D3 500 Feed Grade product was specifically destined for a poultry pre-mixture, or that Stour Bay knew the particular ingredients or formulation of Provimi's poultry pre-mixture into which the product was ultimately combined.

254. Mr Kulkarni, QC's case was that he had no need to prove either of those things:

i) Stour Bay, he said, knew that the product was going to be used was for an animal pre-mixture. That was a "particular purpose" for the purposes of section 14(3) of the SGA, the inclusion of the phrase "whether or not that is a purpose for which such goods are commonly supplied" in the section making clear that a "particular purpose" need not be a special or unusual purpose;

ii) Poultry are animals (and, as Mr Kulkarni, QC observed, there was an image of a chicken on the FBL product label, which belied any suggestion that the product was sold as one that was unsuitable for poultry), and the use of the product to make a poultry pre-mixture was, thus, within the stated purpose. There was, furthermore, no evidence that there was something idiosyncratic about a poultry pre-mixture, as compared to a pre-mixture prepared for other animals. Nor, Mr Kulkarni, QC said, was there anything unusual or atypical in the ingredients that were used in Provimi's own poultry pre-mixture, including choline chloride and other elements that might be aggressive to vitamin D3.

255. So far as this last point is concerned, Mr Kulkarni, QC relied on the factual evidence given by Mr Fiedorow in paragraphs 17 and 18 of his witness statement to the effect that all the ingredients used in Provimi's pre-mixture were industry-standard, in the sense that they conformed to the relevant EU regulations, and were ingredients used by the nutrition industry for production of pre-mixes.
256. Mr Fiedorow maintained his position that the ingredients that Provimi used in its poultry pre-mixture were standard ingredients in cross-examination:

"Q. But it's right, isn't it, that the Provimi pre-mixes include choline chloride and hydrated forms of trace elements which are known to be aggressive to vitamin D3?

A. Provimi premixes contain the same ingredients as all the standard poultry ingredients on the market and all the standard poultry ingredients on the market also contain choline chloride and other ingredients which can influence on vitamins. It's not only vitamin D3 but all of them. However, also just to be completely clear about it, it is standard on the market and is required by right(?) So, any other competitor on the market is using the same formulas for poultry.

Q. Right. Let's take that in stages and perhaps you'll answer my question. Do you accept first that the Provimi premixes include choline chloride?

A. Some of them, yes.

Q. And do they include hydrated forms of trace elements.

A. If you can be more specific, which hydrated forms, because I would like to be specific on the chemicals not on the general statement.

Q. Well, iron salts, copper salts.

A. Which copper salts?

Q. I can't be more precise than that, but you're not accepting are you that there were iron salts or copper salts in your premixes.

A. Some copper salts are standard used in any type of the premixes. They are just salts of the copper in the premixes.

Q. And the use of such ingredients is to be avoided if at all possible and is not ideal, is it?

A. I cannot confirm if it has to be avoided because everybody is using this on the market so I cannot confirm it should be avoided.

…

A. I can only say we are using – we were using standard formulas which are on the market and, to be completely clear, those formulas are very often asked by clients, so it is standard on the market.

…

A. There are thousands of the formulations of the premixes so of course there are premixes where the choline chloride is not used because of the nutrition reasons. However, it is standard to use choline chloride, especially in the poultry premixes. It is a standard on the market that everyone uses."

257. Mr Kulkarni, QC also relied upon the common ground between the experts, reflected in paragraphs 11 and 20 of their Joint Memorandum:^[7]

"11. The premixes prepared by Provimi contained ingredients which are known to be aggressive to vitamin D3, such as choline chloride and hydrated forms of trace elements. Mixing these ingredients with susceptible vitamins is to be avoided if possible, but it is not always possible. DP states in his report that mixing these ingredients is usual; CW pointed out that many feed mills add choline chloride separately to the premix and DP concedes that both circumstances are normal. I.e., it is not ideal to mix aggressive ingredients with vitamin D3 but in some cases it does occur and, in those cases, there is a risk that vitamin D3 might be degraded. In those cases, occurrence of degradation will depend on several factors which may need to occur together depending on the nature of the vitamin D3 preparation.

…

20. The conditions of use in premixes containing aggressive ingredients such as choline chloride and hydrated forms of trace elements was not ideal from the point of view of maintaining vitamin stability but is sometimes unavoidable in practice, particularly where feed mills do not have facilities for separate dosing of choline chloride and/or separate dosing of vitamins and trace elements"

258. Professor Whitehead was asked in cross-examination about the inclusion in pre-mixtures of aggressive substances such as choline chloride:

"Q.In that regard, substances like choline chloride and trace elements are pretty typical in European standard premixtures, aren't they?

A. No. In relation to choline chloride, I would say it's certainly not a recommended procedure and diligent companies, if I can put it that way, prefer to add the choline chloride at the time they're mixing the diet rather than putting it in the premix because it's well known that choline chloride is aggressive to a large number of vitamins. Not just vitamin D, but many of the B vitamins are also sensitive to choline chloride.

Q. So, as I understand your evidence, you say that there are some producers that add choline chloride differently, but there's still enough producers that mix the ingredients at the same time that it can be regarded as a usual or a not unusual way of – sorry a not unusual premixture?

A. It's certainly not best practice.

Q. Because you accept that, in the joint memorandum, there will be mills that won't have the facilities for separate dosing of choline chloride or separate dosing of vitamins and trace elements: yes?

A. Could you repeat the question, please?

Q. Sorry, I'll show you that [paragraph 20 of the Joint Memorandum shown]. So my point was you recognise that there are mills that won't have facilities for separate dosing of choline chloride?

A. Or perhaps are reluctant to have facilities because it involves them in extra work.

Q. But if your product is used in one of those mills, which it very well might be, it's inevitably going to be mixed with choline and trace element?

A. That's correct.

Q. There's a sufficient number of those sorts of mills and producers that it is a usual form of standard European premixture, yes?

A. Yes."

259. Mr Kulkarni, QC also pointed to the fact that the SAQUAL report, a report commissioned by Provimi in 2016 to review the vitamin D3 stability issue arising out of the incorporation of FBL's Vitamin D3 500 Feed Grade product in Provimi's pre-mixture, commented that tests carried out by Provimi into the stability of the FBL product had been done in a model premixture:

"… containing substances that are typically present in European premixes but which are known to cause oxidative stress".

260. The effect of this evidence, taken as a whole, in my judgment is this: whilst it is plain that feed and pre-mixture manufacturing practices vary, and whilst "best practice" may be to add choline chloride at a later stage in the process, it cannot be said to be either particularly unusual or uncommon for pre-mixtures to contain ingredients known to be aggressive to vitamin D3, including choline chloride.
261. This conclusion is consistent with the evidence of Mr Gibbons, who said in the course of his cross-examination that he was not surprised that Provimi was using choline chloride in its pre-mixtures:

"Q.… So right from 2009 onwards Provimi bought choline chloride from you, didn't they.

A. Yes.

Q. So given you knew Provimi produced premixture for animal feed, did it come as a surprise to you that the choline chloride that you were selling them was being used for what they do, which is to make premixture?

A. No. We sell choline chloride to every UK customer we have."

262. Whilst the fact that Provimi's pre-mixture was designed for poultry, and whilst the precise ingredients and formulation of Provimi's poultry pre-mixture, may have been unknown to Stour Bay, there is no evidence that either of those matters rendered Provimi's pre-mixture somehow atypical or unusual compared to other animal or poultry pre-mixtures.

(b) What changed in 2015?

263. Logically, one might expect to deal first with the question of whether there was a breach of the SGA implied terms before moving on to consider whether that breach caused the problems with the poultry that occurred in 2015 and the losses that are alleged to have been suffered by the Claimants as a result.
264. To a large extent, however, Provimi's case as to the alleged unsatisfactory quality and/or unfitness for purpose of the Vitamin D3 500 Feed Grade product, and accordingly its case as to breach, was based on a negative inference from the degradation of the vitamin D3 content that occurred in 2015. It is, therefore, convenient for me to consider first why that happened.
265. The starting point is the recognition and the acceptance by the experts of two matters, namely that:

i) FBL's Vitamin D3 500 Feed Grade product had been supplied in significant quantities and had been used without problem in Provimi's pre-mixture (and also by other customers) in 2013 and 2014; and

ii) Something must have occurred during 2015 which changed the position and led to the problems that arose.

266. This was to some extent reflected in the experts' Joint Memorandum, but it was confirmed in the following exchange at the start of Mr Pickard's cross-examination:

"Q. As you say in your report, an unusual feature of this case is that the FBL Vitamin D3 500 Feed Grade product had been used by Provimi and Cargill without issue from 2013 until the middle of 2015.

A. Yes.

Q. And that was the case notwithstanding that some of the Provimi premixes contained aggressive ingredients including the choline chloride and metal salts -

A. Yes.

Q. - and notwithstanding the fact that the product didn't have a gelatin or other coating and was not spray-dried?

A. Yes.

Q. We know that it had been used by other customers without issue during that time?

A. As I understand it, yes.

Q. And I don't think I need to take you to the document, but let's just see if you recognise the names. Just for people's notes, bundle D tab 153, page 519. There's a purchase confirmation for 3,000 kilograms for Agrofeed and 3,000 kilograms to Premier. Are they names that you recognise?

A. I recognise Premier, yes.

Q. So the issue between you and Professor Whitehead, as I understand your joint memorandum, is what was the cause of the sudden change in the degradation profile or stability of the product in mid-2015?

A. Yes."

267. As Mr Pickard went on to accept later, although the experts had between them originally identified a number of possible reasons for the sudden change in 2015, including an alteration in the formulation of Provimi's poultry pre-mixture and issues at individual Provimi sites, all but two of them had been discounted. The two possibilities that remained were:

i) The heatwave experienced in Europe in late spring/early summer 2015, or

ii) A change (or changes) made in 2015 in the ingredients and/or the process used by FBL to manufacture its Vitamin D3 500 Feed Grade product,

recognising, of course, that those two possibilities might have operated in combination, i.e., that there might have been a change in FBL's manufacturing process in 2015 that made the product less stable or robust, and more likely to suffer degradation by heat, than in previous years.

Temperature

268. So far as the first possibility is concerned, it was common ground between the experts: first, that heat and oxygen are the main methods or causes of vitamin D3 degradation; secondly, that it was unlikely that temperatures above 25°C would have affected the Vitamin D3 500 Feed Grade product while it was in its original packaging; but, thirdly, that elevated temperatures whilst the product was in use in Provimi's pre-mixture would more likely than not affect the stability of vitamin D3.
269. In August 2016, as had been contemplated at the Crevin meeting, Provimi carried out a stability test using a batch of the delivered Vitamin D3 500 Feed Grade product which was incorporated into the Provimi pre-mixture and then maintained at controlled temperatures and humidity settings over a number of days. The study report included the following table of results:

Days	0	8	15	29
Vitamin D3 20°C 60% RH %	100.0	67.2	55.7	43.3
Vitamin D3 25°C 65% RH %	100.0	36.4	31.3	22.5
Vitamin D3 30°C 65% RH %	100.0	24.4	19.7	12.4
Vitamin D3 40°C 65% RH %	100.0	9.8	6.5	3.5

270. As can be seen, the rate of vitamin D3 content degradation markedly increased with temperature: at 20°C the vitamin D3 content dropped from 100% to 67% over a period of 8 days, but even at 29 days remained at 43%; at 40°C, however, the vitamin D3 content dropped from 100% to 9.8% over 8 days, and by the end of 29 days the remaining vitamin D3 content was negligible.
271. Professor Whitehead said that the pre-mixture formulation used for the testing was the harshest possible formulation, but that degradation is something that can always be factored into a pre-mixture production process by increasing the quantity of the vitamin D3-containing ingredient so that, even taking into account predicted degradation, the vitamin D3 content that is left is sufficient for animal nutrition purposes.
272. His main point, however, was that the high temperatures used in the stability test were, he thought, in the same order as the temperatures experienced in the heatwave in Europe in the late spring and early summer of 2015. He maintained the view, reflected in the Joint Memorandum and in his own reports, that it was the European heatwave that was the sole reason why problems were experienced in 2015 but not in prior years.
273. So far as the heatwave is concerned, although neither of the experts was a meteorologist, paragraph 9 of their Joint Memorandum recorded that they both agreed that:

"9. Europe experienced a heatwave in late spring early summer 2015 which would have resulted in goods in transit, in store and in use being exposed to temperatures above 25°C and in some cases above 30°C."

The significance of the 25°C figure is that, as set out in paragraph 119 above, the FBL Product Information Sheet and the product label contained an instruction that the product should be "store[d] in cool place below 25.0 C."

274. Mr Kulkarni, QC suggested in his cross-examination of Professor Whitehead that the temperatures at the location of the Claimants' factories were not particularly high, and were not markedly higher than those in prior years. Ms Ansell, QC complained that, given the agreement between the experts, this line of questioning was impermissible, but I do not consider it was.^[8]
275. The material put by Mr Kulkarni, QC to Professor Whitehead in cross-examination in this context consisted of the following:

i) Two sets of temperature records produced by Provimi in relation to temperatures at its Spanish and Polish facilities:

a) The chart of temperatures at Provimi Spain's Colmenar facility showed temperatures reaching 25°C in July 2015, but temperatures had been almost as high in preceding years;

b) Plots of temperatures at Provimi Poland's Kiszcowo facility appeared to show temperatures above 25°C and up to (but not exceeding) 30°C during the first half of August 2015.

The value of this material was somewhat diminished by the absence of heatmaps or information about the location of the heat sensors at the facilities, although I was told that the Kiszcowo plots showed the position at three different areas;

ii) Temperature graphs produced from an online source (https://www.worldweatheronline.com) showing outside air temperatures at Crevin, France, Colmenar, Spain, Kiszcowo, Poland and Dalton, England, the location of the Claimant companies' facilities. These appeared to show:

a) Temperatures exceeding 30°C at Colmenar during the middle of 2015;

b) Temperatures reaching 30°C at Kiszcowo during the middle of 2015; another set of charts showing temperatures at the nearby Poznan-Lawica airport in Poland showing temperatures exceeding 35°C;

c) Temperatures reaching but not exceeding (the scale of the graphs made it very difficult to tell) 25°C at Crevin in the middle of 2015; the position in previous years did not appear to be materially different; and

d) Temperatures not rising above 20°C at Dalton at any time during 2015. Professor Whitehead said that this was consistent with his recollection that the heatwave was a central European problem, and that it never quite reached the United Kingdom. No claim was made by Provimi UK for an amount paid in settlement of a customer claim.

276. Professor Whitehead was somewhat sceptical as to whether these graphs – which themselves showed high temperatures in Spain and Poland - reflected the conditions inside the factories. He also made the point, consistent with the common ground in paragraph 9 of the Joint Memorandum, that one had to take account of the fact that, once the pre-mixture had been made, it was then transported by lorry to feed mills where it would be exposed to environmental conditions.
277. He expanded on this point in his re-examination: ^[9]

"Q. Could you just expand on what you were going to say because I think you were stopped?

A. Yes. These lorries are not refrigerated. Sometimes the factories can be a considerable distance - the feed mills can be a considerable distance from the premix factory and anything could happen. If the driver stopped for a cup of coffee and leaves the lorry in the sun, temperatures could reach quite high."

As Ms Ansell, QC correctly observed, at least two batches of allegedly defective product produced at Provimi's premises in Crevin (in Brittany, France) ended up in Italy.

278. Mr Pickard did not dismiss high temperatures as a factor in the deterioration of the vitamin D3 content in the pre-mixture, but said that he thought that it was only a contributory factor. This was reflected in paragraph 34 of the Joint Memorandum and confirmed in his cross-examination. During the course of Mr Pickard's cross examination, having put to him that there had been no complaints in relation to the product delivered to Provimi prior to April 2015, there was this exchange:

"Q. … So having looked at those, there does seem to be quite a strong correlation, doesn't there, between date of delivery of this product and problems experienced with this product?

A. There does seem to be a correlation in that the complaints arose during that delivery period.

Q. Yes.

A, That doesn't necessarily mean that that was the causation.

Q. No, but I think you agree that extreme heat is highly likely to have affected the stability of the vitamin D3 in the premixture.

A. Yes, it is likely, yes.

Q. We've seen that before there was that extreme heat - we've looked at two years, four months roughly – no problems at all?

A. Okay, yes.

Q. So let's see if we can at least agree this: it's wrong isn't it, to say that the formulation was not sufficiently robust to withstand typical manufacturing conditions because it clearly did withstand typical manufacturing conditions for two years and four months?

A. Yes, and I think I mentioned that in my report, that that method of manufacture, although quite unusual, I think, it's not necessarily the case that it could not produce a stable product.

Q. No, but I'm asking a different question, if I may. You say that the formulation was not sufficiently robust to withstand typical manufacturing conditions but the point is that it did withstand typical manufacturing conditions for two years and four months.

A. Yes, those particular batches did, yes."

279. For his part, whilst maintaining his opinion that the reason why problems arose in 2015, but not in the prior years, was the extreme heat, Professor Whitehead accepted that, if the Vitamin D3 500 Feed Grade product (or rather the vitamin D3 content within it) had been better protected, the position would likely have been different:

"Q. So I would suggest to you that what those paragraphs [paragraphs 21 and 33 of the Joint Memorandum] we've looked at say is that the effect of elevated temperatures is likely to be felt if the product is not adequately protected, yes?

A. It's likely to be more.

Q. Yes?

A. Yes.

Q. So I must suggest to you that the correct position – I think we can both agree that elevated temperatures can worsen vitamin D3 degradation, but if the product was properly protected, the elevated temperatures would likely not have led to the degradation of the product: yes?

A. Yes, that's reasonable."

Change in manufacturing process

280. As I explained earlier, whilst Professor Whitehead considered that the problems in 2015 arose because of the extreme temperatures, Mr Pickard postulated that the difference in outcome between 2015 and prior years might have been attributable to a change (or changes) in the ingredients and/or the process used by FBL to manufacture its Vitamin D3 500 Feed Grade product.
281. Mr Pickard pointed to two possible changes:

i) A change in the source of cholesterol, the base ingredient for the Vitamin D3 500 Feed Grade product; and

ii) A change in the quantity in which, and the point in time at which, silicon dioxide was added during FBL's manufacturing process.

282. So far as the first is concerned, vitamin D3 is prepared by a chemical synthesis process for which the main starting compound is cholesterol, which is extracted from lanolin that is derived from sheep's wool. Prior to December 2014 FBL had been purchasing cholesterol for its vitamin D3 products from a Chinese supplier, Zhejiang Garden.
283. In December 2015 FBL began to manufacture cholesterol itself. As an email exchange between Prashant Nagre of FBL and Mr Gibbons on 13 May 2015 explained, FBL had experienced some unforeseen problems, which had meant that production had been stopped until April 2015 and so the timing of deliveries by FBL to Stour Bay and by Stour Bay to Provimi had been delayed.
284. This gave rise to a question as to whether the Vitamin D3 500 Feed Grade product supplied to the various Provimi entities from around April 2015 had been produced using Chinese cholesterol or using FBL's own manufactured cholesterol; and, if the latter, whether the change in the source of cholesterol could have had an adverse impact on the stability of the vitamin D3 component.
285. There was no real evidence to address the first part of that question, although both experts expressed views or made assumptions as to whether the email I have just mentioned meant that FBL was more likely to have been using Chinese cholesterol up until April 2015, or whether it was more likely that FBL was trying to use its own manufactured cholesterol during that period.
286. As for the second part of the question, in paragraph 24 of the Joint Memorandum the experts agreed the following:

"We agree that the change from externally sourced cholesterol to internally synthesised material would be most unlikely to affect the intrinsic stability of the resulting vitamin D3 molecule which has 'fixed' chemical properties. However, there is a distinction to be drawn between the stability of the D3 molecule and the stability of the D3 product. As detailed below, we are not agreed as to the potential consequences for the D3 product of the change in the production of cholesterol."

287. The Joint Memorandum went on to record at paragraphs 30 and 31 that, whilst Professor Whitehead's opinion was that the change in cholesterol would not have resulted in a change in the stability of the product, Mr Pickard thought it might have done so:

"31. DP considers that although FBL producing cholesterol in-house was unlikely to affect the stability of the vitamin D3 molecule it may have had an impact on the stability of the vitamin D3 product because some characteristics of the vitamin D3 active ingredient may have been altered. For example, lower potency vitamin D3 from the in-house FBL process compared to that produced from bought-in cholesterol would have resulted in a need to increase loading of active ingredient in the formulation, which may have required formulation adjustments, or may have affected the characteristics of the agglomerated powder. Other factors could also have had an effect; for example, the viscosity of the vitamin D3 resin might have affected the ability of the formulation to bind or combine. These factors may have contributed to a decline in the physico-technical quality of the product and the sudden emergence of the stability problems, but we both agree that in the absence of data this is speculation."

288. The final clause of this paragraph speaks for itself, but Mr Pickard confirmed in his oral evidence that he was not able to say whether FBL's home-produced cholesterol (if it had been used) had lower potency than cholesterol purchased from the Chinese; that he was not able to say whether the use of FBL's home-produced cholesterol gave rise to issues with viscosity; and that he was not able to say whether the use of FBL's home-produced cholesterol, if it had any effect on the product at all (itself a matter of doubt), would have resulted in a change for the better or for the worse so far as stability is concerned.
289. Professor Whitehead was scathing in his oral evidence about the suggestion that the change in the source of the cholesterol might have an adverse effect on the product:

"A. … I have grave reservations about Mr Pickard's contention that changes in the manner of cholesterol production affected anything, let alone the direct evidence that I have that there was no effect.

If cholesterol from either Indian sheep or Chinese sheep ends up as a crystalline product, a pure crystalline product, then the subsequent procedures that Fermenta were carrying out on the Indian-derived cholesterol would have been precisely the same set of procedures that they used on the Chinese derived cholesterol to end up with the vitamin D. So no part of that process should have changed. They would have used the same process.

The only difference in the process was that they were taking lanolin from, let's say, Indian sheep – I have no proof that it was Indian sheep but let's assume it was Indian sheep as opposed to the Chinese presumably, let's say used Chinese sheep – and they arrived at a crystalline product, and its already been pointed out that if you have a pure crystalline product, then you can't tell anything about its source. And from then on, the production process would have been the same so there's no reason to believe that there had been any change in the nature of the final vitamin D product. All these questions about viscosity and what-not is totally irrelevant.

…

Q. But if your in-house cholesterol yields 5% D3, for example – I'm just coming up with these percentages – if your in-house cholesterol yields, say, 5% and the purchased cholesterol yields 10%, you're going to have to use more of the in-house cholesterol and that's going to affect the rest of your formulation.

A. No, the cholesterol is purified to a crystalline form. You're then taking a known quantity of crystalline cholesterol to go through the other processes to end up with your vitamin D, and that process is unchanged. So you started from a pure product, you're using an unchanged process, so what you end up with is the same whether you started off with lanolin from an Indian sheep or from a Chinese sheep.

Q. Mr Pickard also says that the viscosity of the vitamin D3 resin might affect the ability of the formulation to bind or combine. That's possible, isn't it?

A. But why should viscosity change?

Q. The point is, if you're using a different cholesterol, then that has an impact –

A. No, cholesterol – sorry, cholesterol is cholesterol. You have a crystalline product. It doesn't change."

290. The second change referred to by Mr Pickard was a process change concerning the quantity of silicon dioxide added during the process of manufacturing the Vitamin D3 500 Feed Grade product, and, more significantly, the point in time in the manufacturing process at which the silicon dioxide was added.
291. The position in this regard is that on 30 June 2020, in a Response to a Request for Further Information of an allegation in the Defence that no "significant" changes had been made to the manufacturing process for the FBL product between April 2013 and June 2015, Stour Bay said that FBL had since confirmed that "no change" had been made to the manufacturing process at all.
292. It appears that this may not have been quite right, because two flow charts were disclosed, one dated 19 January 2015 and another dated 26 May 2015, each ostensibly showing FBL's manufacturing process as at that date. In the first, silicon dioxide was introduced at a fairly late stage in the process; in the second, it was added near the start as part of the mixing and granulation process.
293. The Joint Memorandum in paragraphs 30 and 32 recorded that Professor Whitehead considered the change to be minor and one that would not result in any change that would affect the stability of the product, but that Mr Pickard's opinion was that the change could affect the character of the resulting agglomeration, although as with the change in the source of cholesterol, this was a matter of speculation:

"32. The flow charts provided by FBL indicate differences in the quantities of silicon dioxide used in the formulation and the order of addition. CW believes that the differences suggested by the flow-charts are minor and likely to have no impact. DP believes the type of silicon dioxide used, the quantities used, and the order of addition are significant elements of the manufacturing process and product composition. This is because silicon dioxide, routinely described as an anti-caking or free-flow agent, is a technological ingredient which is engineered to have many different properties. Typically, the properties varied are the ability to absorb liquids; particle size and shape; 'surface character' with respect to how the ingredient interacts with moisture or fats and oils (hydro-lipo character). Therefore, changes in how silicon dioxide was used may be indicative of a formulator's response to changes in the physico-technical nature of a starting material. In the absence of changes in starting materials, changing the order of addition or quantity of silicon dioxide could affect the character of the resulting agglomeration. Again, both experts agree that in the absence of data this is speculation."

294. In his cross-examination, Mr Pickard explained that there were a number of different types of silicon dioxide, and that, depending upon when they were added, they might have been intended to perform different functions in the manufacturing process - if added earlier, perhaps performing the function of liquid absorption, if added later, perhaps performing an anti-caking function.
295. Mr Pickard acknowledged, in his cross-examination, however, that he had no information about the particular type of silicon dioxide used by FBL, and that all he was able to say was that the change "could" or "may" have had an impact:

"A. Again, I think the silicon dioxide could have affected certain characteristics of the product but we just don't know that. We don't have that information. It could have affected the particle size. It could have affected the flowability."

This was, in fact, the tenor of Mr Pickard's evidence in relation to both his suggested changes:

"Q. Right. So [the deterioration in vitamin D3 content has] either got to be because it was hot – or this is how I understand Professor Whitehead and your – or because there were changes, and I'm putting it to you that those changes simply – there's no evidence that there were such changes that had an impact in June/July 2015.

A. We have information which suggests that there may have been changes and, if those changes happened, then they may have had an impact in June/July 2015"

(emphasis added).

296. In relation to both of the suggested changes in the manufacturing process, Professor Whitehead pointed to an analysis of two batches of the Vitamin D3 500 Feed Grade product, one produced in February 2015 and the other in March 2015 but neither used in the Provimi pre-mixture until June 2015.
297. The analysis of both batches showed large depressions in the initial quantity of vitamin D3 (one of 49% and the other of 90%) similar to the results seen with later produced batches. Professor Whitehead suggested that, as these batches were manufactured prior to April 2015, they must have been manufactured using Chinese cholesterol, and that as they were manufactured prior to 26 May 2015, they must have been manufactured following the process in the 19 January 2015 flow chart.
298. Mr Pickard was asked about this:

"Q. So that seems clear evidence, doesn't it, that the same problems occurred when you had Chinese cholesterol and when you had the different silicon dioxide as it did after the change?

A. It's clear, yes.

Q. Right. So doesn't that mean then that the degradation, the sudden change in 2015, had nothing to do with the changes in manufacture but was all to do with the fact that they were suddenly in too hotter temperatures?

A. It indicates that in some cases some batches, the change with respect to cholesterol apparently did not have an effect, yes.

Q. No, and the same with the silicon dioxide?

A. Again, I think the silicon dioxide could have affected certain characteristics of the product but we don't know that. We don't have information. It could have affected the particle size; it could have affected the flowability."

Conclusions

299. If it were necessary for me to prefer the views of one expert over those of the other in relation to the impact of the suggested changes, I would prefer the evidence of Professor Whitehead. But, in reality, I have no need to do so.
300. Mr Pickard, as he frankly accepted, had been invited to speculate about whether changes in the manufacturing process might possibly have affected the end product, and that was what he had done. But his speculation, unsupported by evidence and undermined by the analysis of these samples, is not a basis upon which I could properly find that it was a change to the manufacturing process that was the reason for the different outcomes with the use of the Vitamin D3 500 Feed Grade product in 2015.
301. As I explained earlier, the experts agreed that, given that the Vitamin D3 500 Feed Grade product had been sold in substantial quantities and had performed satisfactorily in 2013 and 2014, and had done so in a pre-mixture that contained aggressive ingredients, something must have occurred in 2015 to change the position. The experts also agreed that there were only two potential candidates: high temperatures and change(s) in the manufacturing process.
302. In light of the evidence I have heard, and in the absence of any concrete evidence that any change in manufacturing process would have made a difference, I find that the critical difference was the high temperatures, above 25 degrees centigrade, to which the Provimi pre-mixture containing the product was exposed in the late spring and early summer of 2015.
303. In saying that, I recognise – as Professor Whitehead accepted in the passage in his evidence that I set out at paragraph 279 above – that it may be that, if the Vitamin D3 500 Feed Grade product had been protected by a gelatin or other coating, the vitamin D3 content would have survived these higher temperatures. Equally, it may be that, if the Provimi pre-mixture had not contained the aggressive ingredients it did, any additional degradation resulting from high temperatures would not have been sufficient to cause the problems with the poultry that occurred.
304. But, as between the period before April 2015, when FBL's Vitamin D3 500 Feed Grade product was used successfully, and after April 2015 when it was not, I find that it was the use, transportation and storage of the Provimi pre-mixture containing the product in the high temperatures that were experienced in this later period that made the difference.

(c) Breach

305. Against that background, I return to the question of breach of the SGA implied terms. As I have indicated already, this issue arises only if I am wrong about the incorporation of the Stour Bay T&Cs.

Satisfactory quality

306. Section 14(2) of the SGA provides for an implied term that goods supplied under a contract must be of satisfactory quality. Section 14(2A) explains that goods are of satisfactory quality if they meet the standard that a reasonable person would regard as satisfactory.
307. The test is objective, looking at the question from the perspective of a reasonable person in the position of the buyer who is to be attributed with knowledge of all relevant background facts: see Bramhill v Edwards [2004] EWCA Civ 403, [2004] 2 Lloyd's Rep 653 at [39] (Auld LJ). The burden of proof lies on the party asserting that the goods were not of satisfactory quality, here Provimi: see Bramhill at [41].
308. Section 14(2A) indicates that, in deciding whether goods are of satisfactory quality, two particular matters fall to be taken into account: any description of the goods, and the price paid for them (if relevant). Section 14(2B) goes on to explain that the quality of the goods includes their state and condition, and that, in appropriate cases, aspects of quality can include:

i) Fitness "for all the purposes for which goods of the kind in question are commonly supplied"; and

ii) Durability.

309. So far as the first of these two matters is concerned, as was made clear in Balmoral v Borealis, it is important not to overlook the significance of the words "in appropriate cases" and the nature of the product under consideration and the range of possible uses for that product.
310. Balmoral v Borealis was concerned with the sale and purchase of borecene, a type of polyethylene, which was used by the purchaser, by a process known as rotomoulding, to manufacture oil tanks for industrial and domestic use. At [140] Christopher Clarke J said this about section 14(2) of the SGA and its inter-relationship with section 14(3):

"Section 14(2) of the Sale of Goods Act 1979 is primarily directed towards substandard goods. Although there is an overlap between sections 14(2) and (3) the function of 14(2) is to establish a general standard which the goods in question are required to reach, and not to ensure that they attain some higher standard of fitness for a particular purpose made known to the seller. In appropriate cases the question as to whether goods are of satisfactory quality may be determined by considering whether they are fit for all purposes for which goods of the kind in question are commonly supplied: section 14(2B)(a); Jewson Ltd v Leanne Teresa Boyhan [2004] 1 Lloyd's Rep 505. For a material that has a very wide range of possible uses, and which is to be used and transformed by a specialist manufacturer for his own particular purposes, that seems to me somewhat too wide a test, particularly when polyethylene although commonly supplied for oil tanks is not, in respect of some grades, suitable for that purpose."

311. On the facts, Christopher Clarke J he concluded that Balmoral had not established that the borecene was of unsatisfactory quality because it was unsuitable for rotomoulding generally; he said that whether it was suitable for the particular purpose of constructing above ground static tanks to be used for storing oil over long periods fell within the reach of section 14(3).
312. As for the second factor – durability, which is not dissimilar to what was referred to in the present case as robustness – it is important to recognize that this is an inherently relative concept. One product may be better (or differently) made and, as a result, more durable or robust than another; but, bearing in mind, in particular, the price paid for the two products, both may be regarded as of satisfactory quality. Depending upon how, and in what conditions, the purchaser intends to use the product, he or she may be happy to receive a less durable or less robust product, but to pay less.
313. The point was one that Ms Ansell, QC discussed with Mr Pickard in the context of the ability of a Vitamin D3 Feed Grade product to withstand combination in a pre-mixture with aggressive ingredients, in the course of which she pointed out the experts' agreement that the price of the FBL product was 10-20 percent cheaper than the Chinese alternative:

"Q. What I want to just explore with you is you say: 'In my opinion …' I'll just rephrase your opinion as that products can be used with aggressive ingredients but many premixes don't contain such ingredients.

A. Yes, some do, some don't.

Q. Right, but what you go on to say is 'Products which can be demonstrated to do this usually attain a market advantage where potency on farm is a factor which overrides price.'

A. Yes.

Q. So people will pay more, basically, if they've got a product which will work with aggressive ingredients?

A. Yes. When you look at these types of ingredients, you may say, 'We'll, we have to pay a premium for this product. It provides additional technical advantages. Are these technical advantages relevant for this particular market?' They may or may not be and you may then decide which product to use. That's what I mean.

Q. I know, I don't disagree, but I think it also follows from that that it's reasonable for a cheaper product, if I can put it in that way, not to be compatible with aggressive ingredients. So you pay a cheaper price but it's still reasonable because it performs satisfactorily with less aggressive – for all those premixes, which, as you say, don't have the aggressive ingredients included?

A. You could pay a cheaper price. I don't know how the Fermenta product compared in price to other products on the market.

Q. Well, I thought in the memorandum you'd agreed that it was being offered at a price that was 10% to 15% less than other market - shall we have a look –

…

A. Yes, Okay. Sorry, I'd forgotten that point."

314. Although Ms Ansell, QC's point was made by reference to the ability of the Vitamin D3 500 Feed Grade product to withstand combination with other intended ingredients in a pre-mixture, the same point might be made about the ability of a Vitamin D3 500 Feed Grade product to withstand high temperatures without unacceptable degradation of the vitamin D3 content.
315. As I observed earlier, Professor Whitehead explained that one can deal with degradation simply by increasing the quantity of the vitamin D3 product added to a pre-mixture. One product may be able to withstand the temperatures ordinarily encountered, but require higher dosing if exposed to higher temperatures; another product may be manufactured in a way that provides better protection for the vitamin D3 content so that this is not required, but it may be more expensive.
316. In this context, and as section 14(2A) requires "all the … relevant circumstances" to be taken into account, in addition to the price paid for the product the instructions provided with the product will obviously be important. In the present case, in the MSDS, in the FBL Product Information Sheet and on the product label (which was pictured on the information sheet and on the product itself), Provimi was told that:

i) There were incompatibilities between the product and certain other materials, including copper and copper alloys and iron and iron salts (the MSDS);

ii) The product should be stored "in a cool place below 25°C" (the FBL Product Information Sheet and the label) and should be used immediately after opening (the label) or at least within a short period after opening (the MSDS).

These instructions hardly give comfort as to the product's durability or robustness if incorporated into a pre-mixture that contains potentially incompatible materials which is then exposed to temperatures above 25 degrees centigrade, as I found occurred in 2015.

317. Ms Ansell, QC also relied upon section 14(2C) of the SGA, which provides that the implied term of satisfactory quality does not extend to any matter making the quality of goods unsatisfactory that is specifically drawn to the buyer's attention before the contract is made or, where the buyer examines the goods before the contract is made, any matter making the quality of the goods unsatisfactory which that examination ought to reveal.
318. I found earlier that:

i) Provimi was expressly told, both in the 30 April 2010 email to Mr Piccolin and in the documents submitted as part of the approval process, that FBL's Vitamin D3 500 Feed Grade product did not contain gelatin and thus could not have a gelatin coating: see paragraphs 53, 73 and 98-102 above; and

ii) Provimi also knew – or it certainly ought to have known from its examination of the sample of the product and/or from the accompanying documents (Mr van der Eijk having acknowledged the importance of a coating) - that FBL's Vitamin D3 500 Feed Grade product did not, in fact, contain a coating at all: see paragraphs 84 and 98-102 above.

319. My conclusion that the Provimi Gelatin Specification did not form part of the contract would not preclude Provimi from arguing that, whether or not specifically required, a Vitamin D3 500 Feed Grade product that was supplied without a gelatin or other coating was not of satisfactory quality for the purposes of section 14(2). But, in light of the findings referred to in paragraph 318 above, such an argument is not open to it.
320. Even then, Provimi might argue that, accepting that the Vitamin D3 500 Feed Grade product did not have a coating, FBL should have incorporated into the formulation and/or the manufacture of the product some other protection making the vitamin D3 content less susceptible to degradation, for example a different anti-oxidant formulation, and that, without any such alternative protection, the product was not of satisfactory quality.
321. The difficulty Provimi faces, in that regard, however, is that the Vitamin D3 500 Feed Grade product was, in fact, used satisfactorily in its poultry pre-mixture (even though, although not unusually, it contained aggressive ingredients) for more than two years prior to April 2015. There is also the common ground between the experts, as recorded in paragraph 19 of the Joint Memorandum that:

"19. Whilst the product did not meet the requirements of the Provimi Ingredient Specification and was not of a typical type for use in feed it was clearly suitable for use in feed under certain circumstances since Provimi/Cargill successfully used the product for a period without any problems."

322. As Christopher Clarke J said in Balmoral v Borealis, the function of section 14(2) is to establish a general standard which the goods in question are required to reach. In light of the evidence I heard, even if the Stour Bay T&Cs had not been incorporated and thus a case under section 14(2) had been open to Provimi, I would not have found that the Vitamin D3 500 Feed Grade product breached that general standard.

Fitness for purpose

323. Section 14(3) of the SGA requires that, where the particular purpose for which a product is being bought is expressly or implicitly made known to the seller, the product must be reasonably fit for that purpose.
324. A "particular purpose" may be a special or unusual purpose, but it may simply be the ordinary purpose for which a product of the kind in question is used. In such a case, the unfitness of the product for that ordinary purpose may well give rise to a breach of both sections 14(2) and 14(3). For the purposes of section 14(3), however, where the intended purpose is special or unusual one, which can be satisfied only by a product having particular qualities, knowledge of a general purpose may not do.
325. This last point is reflected in the judgment of Lloyd LJ in Aswan Engineering Co. v Lupdine Ltd. [1987] 1 WLR 1. The case concerned plastic pails used to hold a liquid waterproofing compound for shipment to Kuwait, which melted when left stacked in on the quayside in Kuwait in extreme heat. At page 17F-H, in the context of an argument that the pails were unfit for purpose, Lloyd LJ said this:

"Moreover, even if there had been an implied condition, I should have declined to hold that Thurgar Bolle [the manufacturers of the pails] were in breach. In the Hardwick Game Farm case [1969] 2 AC 31, 114-115, Lord Pearce said:

'It was argued that such a purpose was too wide and had not enough particularity to constitute a particular purpose. I do not accept this contention. Almost every purpose is capable of some sub-division, some further and better particulars ... A purpose may be put in wide terms or it may be circumscribed or narrowed … The less circumscribed the purpose, the less circumscribed will be, as a rule, the range of goods which are reasonably fit for such purpose.'

To the same effect is an observation of Lord Wilberforce in the Ashington Piggeries case [1972] A.C. 441, 497, that width of purpose is compensated, from the seller's point of view, by the dilution of his responsibility. If making known that the pails were wanted for export is a particular purpose within section 14(3), as Mr Aikens contends, then the purpose could hardly be wider. A very wide range of goods must be regarded as reasonably fit for that purpose."

326. At page 27E-G Nicholls LJ remarked that, given that the particular purpose communicated was not materially different from the general purpose for which pails were bought, the fitness for purpose condition under section 14(3) added nothing to (what was at that time expressed as) the merchantability condition under section 14(2):

"In my view, however, nothing turns on this because, even if Lupdine were to succeed on this point, their claim based on section 14(3) must still fail. The particular purpose made known was not materially different from, or more precise than, the export purpose stated and considered above in the context of merchantable quality. This being so, in my view, on the facts of this case, if the merchantable quality claim fails, so also does the fitness for purpose claim: the pails were reasonably fit for use in the export trade, even though they were not able to withstand the high temperature of the Gulf when stacked five or six high for several days. Given that the particular purpose made known was not materially different from, or more precise than, the relevant purpose for which the goods were commonly bought, I see nothing surprising in the conclusion that in this case the fitness for purpose condition adds nothing to the merchantable quality condition."

327. It seems to me that the present case is similar to that which Nicholls LJ described. As set out in paragraph 254 above, although Provimi's pleaded case suggested a narrower purpose, the case ultimately pursued by Mr Kulkarni, QC was that the product was required to be reasonably fit for the ordinary purpose of including in an animal pre-mixture, which would include, but would not be limited to, a pre-mixture for poultry.
328. If that is the extent of Provimi's case, it seems to me that the finding that I have made above that the Vitamin D3 500 Feed Grade product was of satisfactory quality for the purposes of section 14(2) of the SGA is essentially determinative of the case made under section 14(3). I am satisfied that the Vitamin D3 500 Feed Grade product supplied by Stour Bay was fit for the general purpose of use in an animal pre-mixture.
329. For the avoidance of doubt, insofar as Provimi seeks to suggest that Stour Bay was required to supply a product that could be used for a poultry pre-mixture that was made with aggressive ingredients and that could be used, without significant degradation of the vitamin D3 content, in temperatures above 25 degrees centigrade, then I reject that case. I do so for two reasons:

i) No such specific purpose was made known by Provimi to Stour Bay, either expressly or implicitly; and/or

ii) In the circumstances, I would not regard it as reasonable for Provimi, which alone knew the particular constituents and formulation of its pre-mixture and the conditions in which it would be used, and which tested and approved the FBL Vitamin D3 500 Feed Grade product in advance, to rely upon the skill or judgment of Stour Bay to supply a product fit for that specific purpose.

330. So far as the second of these reasons is concerned, as the authorities make clear (see Jewson Ltd v Boyhan (PR of the estate of Kelly) [2003] EWCA Civ 1030 at [55] (Clarke LJ)), reliance can be partial: where an ordinary purpose is communicated, the buyer may be entitled to rely upon the seller to ensure that the goods are reasonably fit for that ordinary purpose, but not for some uncommunicated more specific purpose.
331. The following passage in the judgment of Lord Steyn in Slater v Finning Ltd. [1997] AC 473^[10] at 487E-H explaining the closely linked concepts of the communicated purpose and reliance is pertinent:

"The correct approach is well settled. In Goode, Commercial Law, 2nd ed. (1995), p. 335, Professor F Roy Goode explains:

'The seller is entitled to assume that the goods are required for their normal purpose, or one of their normal purposes, unless otherwise indicated by the buyer. Accordingly, if the buyer requires the goods for a non-normal purpose, he must take steps to acquaint the seller of this fact before the contract is made, otherwise the seller, if unaware of the special purpose for which the goods are bought, will not be considered to undertake that they are suitable for that purpose.'

In other words, the implication will normally be that the goods are fit for the purpose for which the goods would ordinarily be used. For example, if a contractor in England buys pipes from a dealer for use in a pipe-laying project the seller would normally assume that the pipes need merely be suitable to withstand conditions in our moderate climate. If the contractor wishes to use the pipes in arctic conditions for a Siberian project, an implied condition that the pipes would be fit to withstand such extreme weather conditions could only be imputed to the seller if the buyer specifically made that purpose known to the seller."

332. Provimi, as it seems to me, was entitled to rely upon Stour Bay to supply a Vitamin D3 500 Feed Grade product that was generally fit for the purpose of use in an animal pre-mixture. I do not consider that Provimi was reasonably entitled to rely upon the skill or judgment of Stour Bay to supply a product that was fit for the particular use and circumstances that eventuated.

(iv) Causation

333. In light of my conclusion that Stour Bay was not in breach of the contracts of sale, the question of whether any such breach caused the problems with the poultry that were experienced in 2015 and the losses claimed by Provimi does not arise.
334. In case the matter goes further, however, I record that, if contrary to my findings:

i) The Provimi Gelatin Specification was incorporated into the contracts of sale (in which case, given the absence of a gelatin coating, it was certainly breached); and/or

ii) Stour Bay was in breach of the SGA implied terms in not supplying a Vitamin D3 500 Feed Grade product which contained a gelatin or other coating

then, consistently with the evidence given by Professor Whitehead set out in paragraph 279 above, I would have found that the absence of such a coating was a least an effective cause of the problems that occurred and the losses that were suffered: see Chitty on Contracts (33rd ed.), paragraph 26-076.

I. Loss and Damage

335. In light of my findings so far, the points that arise under this heading do not arise for consideration, but in deference to the submissions made by the parties I will deal with them, albeit briefly.
336. As I explained in paragraph 13 above, there were two aspects to Provimi's claim:

i) The amount of EUR 2,029,090.12 paid by Provimi France, Provimi Spain and Provimi Poland in settlement of customer claims; and

ii) The refund of EUR 69,413.03 that Provimi said Stour Bay had agreed to pay for product that was returned.

(i) The settlements

337. I am satisfied on the evidence that the customer claims identified in paragraph 169 above were made, and that those claims were settled by the relevant Provimi companies in the amounts claimed.
338. Ms Ansell, QC advanced two points as to why those settlements could not be recovered.
339. The first was that she said that Provimi had failed to mitigate its loss by failing promptly to notify its customers of the problems with the pre-mixture and failing to withdraw the pre-mixture from the market as soon as the vitamin D3 degradation had been discovered.
340. So far as this is concerned, the onus of proof rests on a defendant who claims that a plaintiff ought, acting reasonably, to have taken steps to mitigate and could thereby have avoided part of its loss: see Roper v Johnson (1873) L.R. 8 C.P. 167, confirmed by the House of Lords in Garnac Grain Co. Inc. v H.M.F. Faure & Fairclough Ltd [1968] AC 1130, 1140 (Lord Pearson).
341. If Stour Bay wished to advance a case that Provimi had failed to mitigate its loss, it was, furthermore, bound to plead it: see paragraph C1.3(g) of the Commercial Court Guide. No such case was pleaded by Ms Ansell, QC, and in my judgment, and as Mr Kulkarni, QC submitted, a case that Provimi had failed to mitigate its loss was simply not open to Stour Bay.
342. To be clear, I would have rejected the case even it had been open to Stour Bay to advance it: the problem Provimi faced was an unusual one, and although customer complaints first came in around July 2015 I do not consider that Provimi acted unreasonably in taking the time it did to investigate matters and to carry out a root cause analysis before it withdrew its poultry pre-mixture from sale.
343. Ms Ansell, QC's second point concerned the reasonableness of the settlements. She directed me to the summary of the principles set out in the judgment of Ramsey J in Siemens Building Technologies FE Ltd. v Supershield Ltd [2009] EWHC 927 (TCC) at [80]. In order for Provimi to recover the amounts paid in settlement of the five claims, she submitted that Provimi had to prove that:

i) Stour Bay's breach of contract caused the loss incurred in satisfying the settlement;

ii) The claims made against Provimi were of sufficient strength reasonably to justify a settlement; and

iii) The amounts paid in settlement were reasonable having regard to the strength of the claims, reasonableness in this context meaning that the settlement was, in all the circumstances, within the range of settlements which reasonable people in the position of the settling party might have made.

344. Ms Ansell, QC had not pleaded any positive case as to why these ingredients were not satisfied, but she was entitled to test Provimi's evidence and to submit that Provimi had not met its burden of proof in relation to these three matters. Her focus was on the second, and in particular the third – whether the settlement amounts were reasonable.
345. Her principal issue was with the three settlements made by Provimi France: Nutrea, Fratelli Borello S.r.l and Fanin S.r.l. In relation to Fanin S.r.l, Ms Ansell, QC suggested that Provimi France had compromised its claim for more than double the amount assessed by Provimi's loss adjusters, CDH. In relation to all three claims she asserted that Provimi could have relied upon its own standard terms and conditions either to defeat the claims or as a negotiating point to achieve a lower settlement value.
346. The first point was simply wrong:

i) In its 5 April 2016 report CDH had said that the losses of Fanin S.r.l. and its customers to date were estimated at EUR 273,06.59, but three line items were marked "to be completed" and CDH commented that the figure did not take into account a number of items and could increase significantly;

ii) By the time of its later 2 June 2016 report CDH put Fanin S.r.l.'s claimed losses at EUR 506,062.86, but they said that the losses of two of Fanin S.r.l.'s customers could increase and they assumed that additional losses would be claimed. They recommended allowing between EUR 983,000 and EUR 1,038,000 for the settlement of the two Italian cases – Fratelli Borello S.r.l. and Fanin S.r.l.;

iii) The settlement agreement in relation to Fanin shows that its claim were settled for EUR 350,000.00 plus an additional EUR 200,000.00 subject to certain conditions, so EUR 550,000 in all. The Fratelli Borelli claim was settled for EUR 187,000.00, so the two claims were settled for EUR 737,000.00, well within the loss adjusters' estimate.

347. As far as the second point is concerned, it is plain from the contemporary correspondence that the application of Provimi France's terms and conditions, in particular clause 7 which limited its liability to the replacement of defective product, was considered. Nathalie Masson, a lawyer within Provimi France, sent an email on 24 November 2015, at a fairly early stage after the claims were made, indicating that she thought the terms and conditions might generally apply, although she expressed some doubt over one issue.
348. Her successor, Erik Benard, gave evidence before me. He explained that there was an additional problem, not addressed by Ms Masson's email, but which he had discussed with Ms Masson and also with Provimi France's general management, concerning whether the limitation provision in clause 7 would apply given that the problem with the pre-mixture was latent:

"A. She was right, but [her] opinion on the applicability of the general terms and conditions is strictly limited to the applicability itself. But I think that here the principle that was adopted by Provimi France was to qualify the problem that occurred at the farmers and under which category of liability we were at that time, and in fact it turns out that the deficiency of D3 vitamin may qualify under French law as it is the applicable law, as a hidden defect.

So, in such cases French law usually consider that any limitation of liability clause is not applicable because, yes, the customer is mainly viewed as a consumer without any knowledge. The only, let's say, non-conformity that could be subject to a limitation of liability clause would be apparent defect, which was not the case. So the only exception that is admitted by French court is where, let's say, the true professionals, the vendor and the buyer, are of the same specialty, which was not the case here because Provimi was manufacturer of premixtures and the customers were manufacturers of complete feed. So, to me, this limitation of liability clause do not apply at all, although the conditions [do].

Q. That's all very interesting, but we don't see any of that, do we, in the email from Nathalie Masson? So when I asked you, 'Did she get it wrong?', is you answer now, 'Yes, she got it wrong'?

A. At that stage, yes, she got it wrong, but I think that she further elaborated on the French law, which is quite the same for years and years regarding the case law regarding defect – let's say hidden defects.

Q. And where do we see any of those corrections or this new advice that you're talking about in any of the documents? I don't think any of those have been disclosed in this litigation. Are you aware of that?

A. No, I'm not aware of that specifically and I do not refer to any specific document, but during my discussion with – when I came to the function of let's say Provimi's in-house counsel, I raised the same question and, during my discussions with the general management, of course, I asked the question if we can consider that we are of the same specialty, and the answer was no."

349. Mr Thorne put to Mr Benard that the position set out by Ms Masson in her email, even if not right, reflected an arguable case, and Mr Benard accepted that this was so. As Ms Masson had only addressed the general applicability of Provimi France's terms, and as she had not dealt with the applicability of a limitation provision in the case of a latent defect, that answer did not seem to me to take matters very far.
350. When it was suggested to Mr Benard that the point could have been deployed in order to achieve a reduced settlement, Mr Benard's response was:

"A. Yes, but, you know, this is the game of the settlement. We tried to reduce the amounts claimed as much as possible in order, yes, to have minimum loss."

351. I am satisfied on the basis of Mr Benard's evidence, which was not contradicted by any French law evidence adduced on behalf of Stour Bay, that because the defect in the pre-mixture was latent, clause 7 was unlikely to apply. I am also satisfied that the settlements entered into by Provimi France were reasonable in the sense set out in paragraph 343 above.
352. So far as the settlements entered into by Provimi Poland (with Syl-Drob) and Provimi Spain (with Huevos-Leon), Ms Ansell, QC's submission was simply that Provimi had not adduced sufficient documentary and/or witness evidence to satisfy its burden of proof that the settlements were reasonable.
353. Mr Kulkarni, QC, however, produced as part of his submissions a very detailed "Note as to Details of Settlements" identifying all the relevant documents, which included veterinary reports confirming the customers' losses, and the witness evidence in relation to the conclusion of the settlements. I have read that material, but will not lengthen this judgment by setting it out in detail. I am quite satisfied that Provimi has discharged its burden of proof.

(ii) The refund for returned product

354. In paragraph 19 of the Particulars of Claim, Provimi plead that:

"19. Further, after the problems experienced with the Product, the Claimants returned quantities of defective Product to the Defendant in respect of which the Defendant had agreed to refund the purchase price. To date, contrary to that agreement, no refund has been provided"

The claim is, therefore, brought to enforce the terms of a contract separate from the original contracts of sale whereby Stour Bay agreed to refund returned product (seemingly, whether defective or not).

355. The Particulars of Claim contain no particulars of the alleged contract, contrary to CPR 16 PD, paragraphs 7.3, 7.4 and 7.5. On 18 January 2021, following disclosure, Clydes wrote to Pinsent Masons asking:

"4.5 Further, please point to the documentation and correspondence that proves the Claimants' claim under Paragraph 19 of the Particulars of Claim."

The response from Pinsent Masons in their letter of 11 February 2021 was:

"With respect to paragraph 4.5 of your letter, we confirm that Provimi's evidence with regard to the claim to a refund on returned product will be addressed in Provimi's witness evidence."

No Provimi witness statement, however, addressed this issue.

356. No details of the supposed contract were contained in Mr Kulkarni, QC's written opening submissions. In his closing roadmap, however, he referred to an email exchange between Mr Gibbons and Mr Piccolin on 1 March 2017 in which, following an email from a Provimi finance manager identifying returned product, Mr Piccolin sent an email saying:

"If you remember we spoke about the below last time we met (Eurotier?), you told me you would solved it. Could you please act accordingly and do the necessary for closing this dossier?"

Mr Gibbons responded:

"I am sorry not to have got back to you but our Lawyers have been involved (also it is further complicated by the fact we remit your payments to Fermenta Biotech).

Fermenta are over visiting us in the UK next week so I can discuss again with them then? I will also discuss things again with our Lawyers – technically it is not our Lawyers but our Insurance Company's Lawyers which again complicates things further.

I will get back to you shortly."

357. Plainly, this exchange does not amount to a binding contract whereby Mr Gibbons promised to provide a refund for returned product. Mr Kulkarni, QC invited Mr Gibbons in cross-examination to agree that his previous remark, recorded by Mr Piccolin, that he would "solve it" meant that he had previously agreed to provide such a refund, but Mr Gibbons did not agree.
358. The contract relied upon is insufficiently pleaded, and is not made out on the evidence. This aspect of the claim fails.

J. Disposition

359. For the reasons set out above, Provimi's claim fails and is dismissed.
360. I will hear from counsel in relation to consequential matters, to the extent that they cannot be agreed. I am grateful to all counsel (and to those instructing them) for their detailed and helpful submissions.

Note 1    There was reference in Stour Bay’s Defence to the Provimi Gelatin Free Specification, but no case was pursued by Stour Bay at trial that the parties agreed that this specification should apply to any of the relevant contracts of sale.     [Back]

Note 2    In the case of Child PARS Agreements for Provimi Ireland the specification attached was similar but not identical, and it bore a different PAC Code.    [Back]

Note 3    Most of the invoices were for the Vitamin D3 500 Feed Grade product, but the schedule included several Stour Bay invoices to Provimi UK for different products. It also included two invoices, numbers 6385 (Provimi Poland, 11 May 2015) and 6369 (Provimi UK, 8 April 2015) which were identified in the schedule to the Defence as, and which appear to me to be, invoices relating to Parent PARS Agreement, number 2104003775 under which allegedly defective product was supplied.    [Back]

Note 4    Six of these invoices appear to relate to product ordered under Parent PARS 2014003775, i.e., the Parent PARS Agreement under which sales of allegedly defective product were made.    [Back]

Note 5    I note that in paragraphs 1.8 and 1.10 of their 25 January 2018 letter of instruction to Mr Pickford, Pinsent Masons said – presumably on instructions – that “Cargill combined the Product in the Premix successfully from 2013 and received no complaints from its customers who purchased the Premix from it and Cargill’s customers reported no ill-effects of their poultry clutches. However, Cargill then began to receive complaints from its customers in Spain, Poland, France and the UK from mid-2015” and that “Cargill had purchased the Product since 2013 andissues of D3 deficiency, causing widespread ill-health in chicken clutches,only began occurring in mid-2015” (my emphasis).    [Back]

Note 6    Necessarily, the quantity of product ordered by a particular Provimi company would fall to be deducted from the larger quantity covered by the relevant Parent PARS Agreement. As a matter of legal analysis, this would seem to involve the partial novation of the initial contract and/or its substitution and replacement by contracts for smaller quantities made between Stour Bay and the individual Provimi companies.     [Back]

Note 7    In paragraph 22 of his supplementary report, Mr Pickard’s qualified his comment that mixing choline chloride and/or other aggressive ingredients was “usual”, saying this: “Many premixture formulations do not contain choline chloride or aggressive forms of trace elements, but equally, it is not unusual for pre-mixtures to contain these ingredients.”     [Back]

Note 8    Ms Ansell, QC told me that, in light of the common ground about the heatwave contained in the experts’ Joint Memorandum, Appendix 1 to Professor Whitehead’s report had not been included in the bundles. In Appendix 1 Professor Whitehead exhibited literature, including an 8 July 2015 article byErdman, which referred to near record temperatures being recorded in Spain (up to 45 degrees centigrade), Germany (up to 40.3 degrees centigrade), France (up to 39.7 degrees centigrade) and Poland (up to 36 degrees centigrade) in late June and early July 2015.    [Back]

Note 9    Professor Whitehead’s comment that pre-mixture was transported to customers in lorries that are not temperature controlled is consistent with the factual evidence: see, e.g., the statement of Mr Martinez at paragraph 9. There was no evidence that Provimi used labels or issued instructions to its customers to store the pre-mixture once delivered at temperatures below 25 degrees centigrade.    [Back]

Note 10    The case was discussed in BSS Group Plc v Makers (UK) Limited [2011] EWCA Civ 809, which both parties cited in supplementary submissions sent to me following the hearing.    [Back]