Coca-Cola Company v PepsiCo Inc

PEPSICO AUSTRALIA HOLDINGS PTY LIMITED ACN 079 719 743

JUDGE: BESANKO J DATE: 28 november 2014 PLACE: ADELAIDE via video link to melbourne

INTRODUCTION 1 This is an application by The Coca-Cola Company ("TCCC") for declarations, injunctions, damages, and other relief against PepsiCo Inc ("PepsiCo"), PepsiCo Australia Holdings Pty Limited ("PepsiCo Australia"), and Schweppes Australia Pty Ltd ("Schweppes"). A judge of this Court ordered that issues of liability be tried prior to, and separately from, issues of quantum. For the reasons that follow, TCCC's application should be dismissed. 2 TCCC, which is incorporated according to the laws of the United States of America, and its licensees are the worldwide manufacturers of the well-known Coca-Cola beverages. TCCC sells these beverages in glass bottles, plastic bottles, and cans. TCCC's glass bottle is well-known and is referred to by TCCC and in the market as the "Contour Bottle". It is common ground that the Coca-Cola beverage has been widely sold in a variety of glass Contour Bottles since 1938 and that the trade marks "COCA-COLA" and "COKE" are well-known in Australia in relation to non-alcoholic beverages. 3 PepsiCo is incorporated according to the laws of the United States and it is the ultimate holding company of PepsiCo Australia, a company incorporated according to the laws of Australia. Schweppes is also incorporated according to the laws of Australia and it manufactures and sells PepsiCo's products in Australia. It is common ground that PepsiCo and PepsiCo Australia have both directed or procured, and had close involvement in, Schweppes' manufacture and sale of the Pepsi products. Furthermore, it is common ground that PepsiCo, PepsiCo Australia, and Schweppes, and each of them have, in the course of trade in Australia, dealt with or provided their existing PEPSI® and PEPSI MAX® branded cola beverages in a 300 ml glass bottle, and that the trade marks "PEPSI®" and "PEPSI MAX®" are well-known in Australia in relation to non-alcoholic beverages. Finally, it is common ground that PEPSI® and PEPSI MAX® branded cola beverages, as sold in the 300 ml Pepsi glass bottle, are non-alcoholic beverages falling within class 32 under the Trade Marks Act 1995 (Cth) and the Trade Marks Regulations 1995 (Cth). 4 TCCC owns four registered trade marks under the Trade Marks Act which are in issue in this proceeding. The details of the marks are as follows. 5 First, TCCC is and has been at all material times the registered owner of Australian Trade Mark No. 63697 ("TM 63697"), which has been registered with effect from 24 April 1937 in class 32 in respect of "Beverages and syrups for the manufacture of such beverages". The trade mark consists of the following sign: 6 Secondly, TCCC is and has been at all material times the registered owner of Australian Trade Mark No. 767355 ("TM 767355"), which has been registered with effect from 14 July 1998 in class 32 in respect of "Beverages in class 32 excluding beers but including non-alcoholic drinks and beverages, mineral and aerated waters; soda water, dry ginger ale, tonic water, lemon squash, bitter lemon, lemonade, orangeade; fruit juices, fruit juice drinks and beverages containing fruit juice and fruit juice flavouring, including mineral and aerated water containing fruit juice or fruit juice flavouring; sports drinks including electrolyte replacement beverages for sports; dietetic and low calorie forms of all the foregoing goods; carbonated and non-carbonated soft drinks and concentrates for making same including beverages in containers; concentrates, syrups, powders and other preparations and substances for making all the foregoing goods". The trade mark consists of the shape of a bottle as shown in the following representations: 7 Thirdly, TCCC is and has been at all material times the registered owner of Australian Trade Mark No. 1160893 ("TM 1160893"), which has been registered with effect from 12 February 2007 in class 32 in respect of "Non-alcoholic beverages; drinking waters, flavoured waters, mineral and aerated waters; soft drinks, energy drinks and sports drinks; fruit drinks and juices; syrups, concentrates and powders for making beverages including syrups, concentrates and powders for making mineral and aerated waters, soft drinks, energy drinks, sports drinks, fruit drinks and juices". The trade mark consists of the following sign: 8 Finally, TCCC is and has been at all material times the registered owner of Australian Trade Mark No. 1160894 ("TM 1160894"), which has been registered with effect from 12 February 2007 in class 32 in respect of "Non-alcoholic beverages; drinking waters, flavoured waters, mineral and aerated waters; soft drinks, energy drinks and sports drinks; fruit drinks and juices; syrups, concentrates and powders for making beverages including syrups, concentrates and powders for making mineral and aerated waters, soft drinks, energy drinks, sports drinks, fruit drinks and juices". The trade mark consists of the following sign: 9 TCCC claims that each of the four trade marks comprise depictions of the Contour Bottle. TCCC claims that its Contour Bottle has a characteristic shape and silhouette, which it describes as a pinch in the bottom portion of the bottle, a relatively wide portion of the bottle above the pinch, and tapering of the bottle from the wider portion up to the neck of the bottle. Photographs of the Contour Bottle as shown in the Statement of Claim are as follows: The Contour Bottle shown in the photograph contains the regular Coca-Cola beverage. There are other forms of the beverage which have reduced sugar or no sugar. The Contour Bottle containing the regular beverage is sold with a red cap or top, whereas the reduced or no sugar variations have different coloured caps or tops, including, with respect to the no sugar variant, a black coloured cap or top. The words "Coca-Cola" on the bottle are written in what is known as the Spencerian script. 10 TCCC claims that, in its advertising and marketing of the Coca-Cola beverage, it has made extensive and repeated use of the Contour Bottle or the silhouette of the Contour Bottle. 11 TCCC claims that the respondents have, in the course of trade in Australia, dealt with or provided cola beverage products in glass bottles which have a pinch in the bottom portion of the bottle, a relatively wide portion of the bottle above the pinch, and a tapering of the bottle from the wider portion up to the neck of the bottle. Photographs of the alleged infringing bottle as shown in the Statement of Claim are as follows: The respondents refer to their bottle as the "Carolina Bottle", and I will also use that description. The respondents also produce regular and reduced or no sugar beverages. The Carolina Bottle shown in the top two photographs contain the regular beverage. The bottle has a blue cap or top. The circular mark above the word "pepsi" contains the colours red, white, and blue, and was referred to as the "globe device". The Carolina Bottle shown in the bottom two photographs contains one of the sugar variations of the Pepsi beverage and it has a black cap or top. 12 Various colour photographs of the Contour Bottle and the Carolina Bottle containing the regular beverages were put before the Court. Unfortunately, these photographs did not include a comparison of the bottles of the same height (i.e., Contour Bottle 330 ml and Carolina Bottle 300 ml). Nevertheless, the following photograph will give the reader a general idea of the colouring of each bottle: 13 TCCC alleges three causes of action against the respondents. First, it alleges that the Carolina Bottle infringes one or more of its four trade marks within s 120(1) of the Trade Marks Act. On the evidence, and as the case was argued, the question of trade mark infringement raises two issues. The first issue is whether the respondents have used the shape of their Carolina Bottle or, in the alternative, that part of the shape of the bottle which comprises the outline or silhouette, as a trade mark within the meaning of ss 17 and 120(1) of the Trade Marks Act. The second issue only arises if the answer to the first issue is in the affirmative. It is whether the shape of the Carolina Bottle, or, in the alternative, the outline or silhouette of the bottle, is deceptively similar to one or more of TCCC's four trade marks. 14 The second and third causes of action which TCCC alleges against the respondents are the tort of passing off and contraventions of ss 52 and 53(c) of the Trade Practices Act 1974 (Cth). TCCC commenced this proceeding on 14 October 2010. That was before the commencement of item 7 in Schedule 7 of the Trade Practices Amendment (Australian Consumer Law) Act (No 2) 2010 (Cth) and, therefore, the Trade Practices Act continues to apply to this proceeding. However, TCCC's application for injunctions is taken to be proceedings for an injunction under s 232 of the Australian Consumer Law (Sch 2 of the Competition and Consumer Act 2010 (Cth)). TCCC claims that it had developed in Australia substantial goodwill and a valuable reputation in the Contour Bottle and in the signs comprising the four trade marks. It claims that its business and goods have become "widely and favourably known and identified in the minds of consumers throughout Australia by reference to the Contour Bottle and the signs" comprising the four trade marks. TCCC claims that members of the public in Australia expect to be dealing with it, or persons authorised or approved by or otherwise associated with it, when they are offered for sale the Pepsi beverages in the Carolina Bottle. TCCC claims that the offering for sale and sale of the Pepsi beverages in the Carolina Bottle constitutes the passing off of the Pepsi beverages as products of TCCC or licensed by TCCC. Furthermore, it claims that such acts constitute a contravention of s 52 and a contravention of s 53(c) of the Trade Practices Act.

WITNESSES 15 TCCC adduced evidence from six witnesses. 16 Ms Lynette Davis is the Group Manager of Marketing Strategy and Insights at Coca-Cola South Pacific Pty Ltd ("CCSP"). CCSP is a wholly-owned subsidiary of TCCC and it is authorised by TCCC to manage the marketing, advertising, and sales promotion of products sold under the various trade marks owned by TCCC in the South Pacific region, including Australia. Ms Merle Le is the Manager - Shopper Merchandising and Marketing in the Customer and Commercial Department of CCSP. Ms Davis and Ms Le gave evidence about the history of TCCC and the Contour Bottle, the advertising and promotion of the Contour Bottle, and the trade channels through which it is sold. The respondents did not criticise the evidence of either Ms Davis or Ms Le. They were straightforward witnesses and I accept their evidence. 17 Ms Kristine Van Ruiten is a private investigator. She produced a number of photographs of bottles and cans displayed in fridges, coolers, and on shelves in different retail outlets. Her affidavit was tendered and she was not required for cross-examination by the respondents. Ms Rachel Peterson is a legal practitioner. She deposed to the circumstances in which TCCC and CCSP became aware of the fact that the Carolina Bottle was being offered for sale and sold in Australia. Her affidavit was tendered and she was not required for cross-examination. I accept the evidence of Ms Van Ruiten and Ms Peterson. 18 Dr Brian Gibbs is Principal Fellow in Marketing and Behavioural Science at the Melbourne Business School of the University of Melbourne. He swore two affidavits dated 16 July 2013 and 13 January 2014 respectively. Dr Gibbs gave expert evidence directed to the conclusions that a consumer would draw from the Carolina Bottle and its relationship to the Contour Bottle and TCCC. It was primarily relevant to the causes of action for passing off and contraventions of the Trade Practices Act. The respondents objected to the receipt of Dr Gibbs' main report, but I overruled that objection for reasons set out below. Dr Gibbs' evidence was confusing and unhelpful and, for reasons set out below, I give it no weight, except where it accords with evidence which I do accept. 19 Mr Barry Crone is an industrial designer with over 40 years' experience in industrial design. He gave expert evidence about the matters that a designer would take into account in designing a bottle, the significant design features of the Contour Bottle and the extent to which those features are reflected in TCCC's trade marks, and the similarities and differences between the Contour Bottle and the Carolina Bottle. He was a straightforward witness who expressed his opinions clearly. However, for reasons set out below, I do not accept a number of his opinions. 20 The respondents adduced evidence from eight witnesses. 21 Mr Robert Cumming is employed by PepsiCo Australia as Beverages Technical Manager. He gave evidence relating to the introduction of the Carolina Bottle into the Australian market. He was a straightforward witness and I accept his evidence. Mr Gary So is a Senior Brand Director of PepsiCo Global Beverages Group, which is part of PepsiCo. He gave evidence about the Pepsi business, its use of trade marks, and its advertising campaigns and sponsorship deals. In cross-examination, he was, at times, very cautious and not inclined to concede anything. However, as it happened, his evidence was, for the most part, uncontroversial and I accept it. 22 Mr Robert Le Bras-Brown was an employee of PepsiCo between June 2003 and November 2010. He gave evidence by video link from the United States. At the time of the events to which his evidence related, he was Director, Packaging Innovation and Development of PepsiCo International, and he played an important role in the design of the Carolina Bottle. It was suggested to him that, in designing the Carolina Bottle, PepsiCo had adopted the contoured shape of the Contour Bottle. I have considered his evidence very carefully. For reasons I will give, I have accepted the substance of his evidence. Mr Bradley Van Dijk is the Managing Director of PepsiCo Australia and New Zealand Beverages at PepsiCo Australia. He gave evidence about the Pepsi business in Australia, the PepsiCo products, and the outlets in which they are sold. He was a straightforward witness and I accept his evidence. 23 Ms Jennifer Evans is the Consumer Affairs Manager at PepsiCo Australia. She gave evidence about whether there had been any complaints made about the Carolina Bottle to PepsiCo Australia. Her affidavit was tendered and she was not required for cross-examination by TCCC. The same occurred in relation to Ms Sarah Vo, who is the Consumer and Quality Relations Manager at Schweppes. She gave evidence about whether there had been any complaints made about the Carolina Bottle to Schweppes. I accept the evidence of Ms Evans and Ms Vo. 24 Professor Jill Klein is a Professor of Marketing at the Melbourne Business School, Melbourne University. She gave expert evidence which commented on the opinions of Dr Gibbs. She was a straightforward witness and her evidence was clear and concise. For the most part, I accept her opinions. Mr William Hunter conducts his own consultancy business involving engineering designs, assistance in product development, and advice as an engineering design expert. His evidence included comments on the opinions expressed by Mr Crone. He was a straightforward witness. I do not accept everything he said, but I do accept the majority of the opinions he expressed.

TCCC AND THE CONTOUR BOTTLE 25 As I have said, TCCC's case is that it has developed substantial goodwill and reputation in the Contour Bottle in Australia and in the signs which are the trade marks. Furthermore, its case is that, in Australia, its business and goods have become widely and favourably known and identified in the minds of consumers by reference to the Contour Bottle. It adduced evidence, primarily from Ms Davis and Ms Le, to establish a factual basis for those propositions. 26 In 1886, Mr J S Pemberton, a pharmacist, created a syrup beverage in Atlanta in the United States. The name "Coca-Cola" was chosen for the beverage and a Spencerian script adopted for use in advertising of the beverage. In 1892, TCCC was incorporated in Georgia in the United States. In 1916, TCCC adopted the Contour Bottle for its Coca-Cola beverage and, at about that time, it was patented in the United States. 27 From 1916, when TCCC adopted the Contour Bottle as the standard bottle for the Coca-Cola beverage and until 1955, throughout the world the 6½ ounce Contour Bottle was the only container used for the sale of the Coca-Cola beverage. In 1955, TCCC introduced larger sized Contour Bottles for the Coca-Cola beverage. In 1993, TCCC introduced into the market in the United States a 20 ounce PET bottle which was based on the Contour Bottle. PET refers to polyethylene terephthalate, which is a material commonly used for plastic beverage bottles, particularly for carbonated soft drinks. 28 In Australia, the Coca-Cola beverage has been sold in the Contour Bottle since 1937. In late 1937, TCCC commenced small scale bottling of the Coca-Cola beverage in Australia in North Sydney. In 1938, the first Coca-Cola bottling plant was set up in Waterloo in Sydney, where the 6½ ounce Contour Bottle was produced. In 1939, additional bottling operations were set up in South Melbourne, in Brisbane and in Adelaide. A further bottling plant was established in Perth during the Second World War. Over time, TCCC granted franchises in the major capital cities and in Newcastle, New South Wales. From 1963, the Coca-Cola beverage was sold in cans as well as bottles. 29 The Contour Bottle came in various sizes, including the 6½ ounce bottle and a 10 ounce bottle. In the 1970s, a 1 litre bottle was introduced. Other sizes to be produced have included a 237 ml bottle and a 185 ml bottle. For a short time, there was even an aluminium bottle based on the Contour Bottle. 30 The Coca-Cola beverages in the Contour Bottle are currently available in Australia in the following sizes: 175 ml, 250 ml, 330 ml, and 385 ml. Each of these bottles has been in continuous use from at least a date Ms Davis was able to identify by reference to archival material available to her (175 ml since 2006, 250 ml since 2003, 330 ml since 2001, and 385 ml since 2006). 31 In 1983, the Diet Coke beverage, which contains artificial sweeteners rather than sugar, was introduced into Australia. It is also sold in 175 ml, 250 ml, 330 ml, and 385 ml Contour Bottles. In January 2006, the Coke Zero beverage, which is a low kilojoule variety of the Coca-Cola beverage that contains no sugar, was introduced into Australia. It is also sold in 175 ml (2006 - 2009), 250 ml, 330 ml, and 385 ml Contour Bottles. 32 The Coca-Cola beverages in PET bottles, which TCCC claims were based on the Contour Bottle, are sold in Australia in the following sizes: 300 ml, 390 ml, 450 ml, 600 ml, 1.25 litre, 1.5 litre, and 2 litre. 33 The dominant colours of the regular Coca-Cola beverage are red and white, the dominant colours of the Diet Coke beverage are silver and white, and the dominant colours of the Coke Zero beverage are black and white. 34 At the present time, TCCC uses four primary trade marks to promote the Coca-Cola beverages. These are the word marks COCA-COLA and COKE, the Contour Bottle, and a ribbon device displayed on PET bottles and cans containing the Coca-Cola beverages. The Coca-Cola beverages are sold in three principal forms, being the traditional or regular formulation and two lower calorie formulations. The lower calorie formulations are promoted under the trade marks DIET COKE (known as COKE LIGHT in some markets) and COKE ZERO. 35 It is convenient at this point to mention that TCCC's "Brand Identity and Design Standards" set out "Defining Elements" and "Supporting Elements" of the Contour Bottle. There are six Defining Elements and they are as follows: (1) Contoured Shoulder; (2) Flutes; (3) Scalloped Transitions; (4) Pinch Waist; (5) Champagne Base; and (6) Overall Proportions (silhouette) (relationship of elements to each other, from base to pinch waist to label area to shoulder). 36 Ms Davis said, and I accept, that, except for a relatively rare case where the bottle is sold in a plastic sleeve masking those features, the Contour Bottle always has the above features. 37 Between the years 2008 to 2012, approximately 40% of the carbonated soft drink market in Australia was comprised of Coca-Cola beverages and, in the same period, the market share of Pepsi for its cola products (Pepsi and Pepsi Max combined) was between about 10 to 12%. 38 The Coca-Cola beverages in the Contour Bottle have been sold in Australia in significant volumes. The Coca-Cola beverages in the Contour Bottle are sold through a variety of trade channels and Ms Davis provided the following estimates of the percentage of all Contour Bottles sold through each channel: the convenience and petroleum channel (10%), the grocery channel (15%), the immediate consumption channel (35%), and the hotels, restaurants and cafes channel ("HORECA") (30%). 39 I have already referred to the role of CCSP. Coca-Cola Amatil (Aust) Pty Ltd, a subsidiary of Coca-Cola Amatil Limited, is currently the exclusive Australian bottler for TCCC in Australia. 40 From time to time, Coca-Cola Amatil (Aust) Pty Ltd, with the approval of TCCC, produces and sells in Australia what it calls "Limited Edition" or "Commemorative" Contour Bottles. For the 2003 Rugby World Cup, a 330 ml Contour Bottle in limited edition packs was produced and sold. As part of a Christmas 2006 promotion, two 330 ml Contour Bottles in four packs were produced and sold. For the 2008 Rugby League State of Origin Grand Final, two 385 ml Contour Bottles were produced and sold in Queensland and New South Wales exclusively in Caltex stores. Finally, in 2008, and as part of a promotion of the Rugby League team Saint George Illawarra, a special Contour Bottle, available only to consumers who signed up as members of the club in 2008, was produced and sold. 41 Coca-Cola products are heavily advertised in Australia. CCSP and Coca-Cola Amatil (Aust) Pty Ltd work together in formulating these advertising campaigns. CCSP conducts approximately five to six national advertising campaigns each year and each campaign can cost between $3 and $5 million and, on occasions, more than that. 42 These advertising campaigns focus on the Coca-Cola brand and its symbols, which include the Spencerian script, the Contour Bottle, the colour red, and the dynamic ribbon device, and they also focus on an association of the Coca-Cola brand with ideas of good times, happiness, fun, family, and friends. 43 Ms Davis gave evidence of various advertising campaigns which she said were examples of the type of campaigns which had been conducted. 44 In 2005, CCSP conducted an extensive advertising campaign in Australia to launch a new flavour of the Coca-Cola beverage known as "Coke Lime". The campaign included outdoor advertising, television, radio and on-line advertising, as well as prominent in-store point of sale advertising and product placement. A black silhouette of a bottle (among other colours) was used during this campaign. In some of the material, it seems to accord fairly closely with TM 1160894, but in other material it seems to have a different cap shape or, more importantly, a different contour shape. 45 The "Coke Side of Life" campaign was conducted between 2006 and 2008 and the campaign involved outdoor advertising on billboards, on the back of buses, on vending machines, on vehicles, in and around stores, on television, on-line, in magazines, and in cinemas. In addition, there was advertising signage at railway stations. The image of the Contour Bottle was a prominent part of this campaign. I do not need to refer to all of the material produced by Ms Davis. It shows that key features of the Coke Side of Life campaign were the Contour Bottle, the Spencerian script, and the colour red. The expenditure on the campaign was very substantial and, by way of example, Ms Davis said that approximately $3.2 million was spent on advertising in the three month period from January to March 2007. 46 CCSP has conducted other campaigns in Australia of a similar nature to the Coke Side of Life campaign. The Contour Bottle is a prominent feature of the advertising in these campaigns. They include the "Open Happiness" campaign in 2009 and campaigns during Christmas, Easter, the summer holiday period (for example, the "Summer of Us" campaign for the summer of 2007/2008), and the football seasons for the Australian Football League ("AFL") and the National Rugby League ("NRL"). 47 Ms Davis accepted that, with some exceptions, it was not until the "Open Summer" campaign in 2009/2010 that a silhouette image of the Contour Bottle appeared without any writing on it or other features. The exceptions were the Coke Lime campaign in 2005, an AFL football shaped as a Contour Bottle in 2005, and cans produced as part of the Summer of Us campaign in early 2008. Ms Davis said that in 2009 there was a directive from the headquarters of TCCC to the effect that the unadorned silhouette was to be emphasised in advertising. The significance of the date upon which the unadorned outline or silhouette of the Contour Bottle was emphasised in advertising will become clear when I come to deal with TCCC's passing off and Trade Practices Act claims. 48 Ms Le provided further evidence of CCSP's approach to the marketing of Coca-Cola beverages. It is very sophisticated. Ms Le said that CCSP divides the path to purchase of a consumer into four zones - the proximity zone, the transition zone, the impulse zone, and the destination zone - and formulates an advertising strategy for each zone. The proximity zone is about 600 m from a retail outlet, and the destination zone is at the primary beverage location within a store and is either a shelf or a refrigerator where the products are available for purchase. Packaging is a key component in TCCC's communication to customers at the destination zone, and, furthermore, distinctive TCCC packaging often appears in the transition zone and the impulse zone so as to "resonate" with products available at the destination zone. 49 Ms Le confirmed Ms Davis' evidence that TCCC uses four primary trade marks to promote the Coca-Cola beverages in Australia (and throughout the world) and they are the word marks COCA-COLA and COKE, the Contour Bottle, and the dynamic ribbon device. 50 Ms Le said, and I accept, that the Contour Bottle, being a glass bottle, is the premium packaging for the Coca-Cola beverages in Australia. That was accepted by all of the witnesses who were asked to comment on the issue. 51 Ms Le gave evidence of the trade channels through which the Coca-Cola beverages, whether in the Contour Bottle or otherwise, are sold. Her categories were the same as Ms Davis, except that she added a fifth category of quick service restaurants. She also referred to a vending channel (i.e., vending machines). She said that examples of the outlets in the convenience and petroleum channel are convenience stores, such as 7-Eleven stores, and petroleum outlets, such as those operated by BP and Shell. Examples of the grocery channel are supermarkets, such as Coles, Woolworths, and independent supermarkets, as well as large retail stores, such as K-Mart and Big W. Examples of the outlets in immediate consumption channel are corner stores and milk bars, fish and chip shops, pizza, and takeaway chicken outlets. She said that the HORECA channel included bistros, coffee shops, hotels, and restaurants, and quick service restaurants included chains such as McDonalds and Hungry Jacks. The Coca-Cola beverages in the Contour Bottle are principally sold through the convenience and petroleum channel, the immediate consumption channel, and the HORECA channel. 52 The Coca-Cola beverages are currently sold in the Contour Bottle in sizes which are designed to be consumed immediately or soon after purchase. Coca-Cola beverages sold in the Contour Bottle are typically refrigerated. They are kept in refrigerators or open air coolers. The refrigerators are enclosed, or at least partly enclosed. 53 Again, like Ms Davis, Ms Le gave evidence of advertising campaigns involving the use of the Contour Bottle as follows: Ruby League State of Origin series between Queensland and New South Wales (mid 2009); "Open Summer" campaign (summer 2009/2010); Easter campaign (Easter 2010); FIFA World Cup (April 2010); and AFL and NRL Final Series (winter/spring 2010). 54 Ms Le described an incident in late 2010 or early 2011 when she saw one of the respondents' products on display in a refrigerator. Initially, she mistakenly believed it was a Coca-Cola beverage in the Contour Bottle until she was close enough to see the blue cap on the bottle. At that point, she realised that it was not a Coca-Cola beverage. 55 The evidence of Ms Davis and Ms Le dealt with two important matters in terms of the issues in this case. 56 The first matter is relevant to the context in which the Contour Bottle is offered for sale and sold. Ms Davis offered some percentage estimates with respect to the trade channels through which the Contour Bottle is sold. The evidence was fairly general and I do not think I can, or indeed need to, make a finding in precise terms. I find that substantial sales of the Contour Bottle are effected in circumstances where a consumer selects the beverage that he or she wants from a refrigerator or open air cooler. 57 The second matter relates to TCCC's advertising of the Contour Bottle and this is relevant to TCCC's reputation in the Contour Bottle. Due to the issues in this case, in examining the evidence of TCCC's advertising, I considered the circumstances in which the whole of the Contour Bottle was shown, including, for example, the vertical fluting, and circumstances where only the outline or silhouette of the Contour Bottle was shown. I also considered the extent to which the Contour Bottle, either its whole shape or only its outline or silhouette, appeared with other signs such as word marks, and the extent to which the Contour Bottle, or only its outline or silhouette, was the prominent feature in the advertisement. 58 Furthermore, in considering this evidence, there is another issue to be borne in mind. There is a dispute between the parties as to the date upon which TCCC's reputation in the Contour Bottle is to be assessed. The respondents submitted that the relevant date is August 2007 and that TCCC's advertising after this date is irrelevant. TCCC submitted that the relevant date is February 2009 and that I can take into account TCCC's advertising up until that date. I deal later in these reasons with the issue of the correct date. 59 The Contour Bottle with all its features, including, for example, the vertical fluting, was featured in TCCC's Australian advertising prior to August 2007, together with the word marks, Spencerian script, and the colour red. Some of the advertising prior to August 2007 involved other signs, but with only the outline or silhouette of the Contour Bottle. Prior to August 2007, the circumstances in which the outline or silhouette of the Contour Bottle dominated the advertising were isolated and fairly rare. The most notable occasion upon which this occurred was the "Coke Lime" campaign in 2005 and, even in the case of that campaign, a somewhat imprecise silhouette of the Contour Bottle was used. Circumstances where only the outline or silhouette of the Contour Bottle was used, such as with the AFL football, were so rare that I do not think that they can form a basis for any conclusion. In 2009, TCCC decided to emphasise the unadorned silhouette in its advertising and issued a directive to that effect. I am not able to detect any significant change in TCCC's advertising between August 2007 and February 2009, and, as I will explain later in these reasons, I do not think TCCC's reputation in the Contour Bottle is any different as between those two dates. 60 Ms Van Ruiten was asked by TCCC to make investigations into single serve non-alcoholic carbonated beverages in Australia. She was instructed to investigate the range of different bottles in which such beverages were packaged and to purchase a broad sample of such products, having particular regard to bottles made from glass and bottles having a waisted or tapered side wall section. She carried out those instructions in February 2013 and purchased 17 bottles (plus the respondents' Pepsi Max beverage), and these 17 bottles were referred to in evidence as the "third party" bottles. Three of these bottles may be described as having waists, although they are different in shape and degree. Those bottles are the Lucozade, Schweppes and PET bottles. However, other than the Schweppes bottle, which was on the market before August 2007, there was no evidence before me as to when the third party bottles first came onto the market. 61 The respondents put in evidence a number of other bottles and I will refer to these as the "additional bottles". All of these additional bottles clearly had waists of one sort or another. However, the evidence went no further than establishing that three of these bottles - a Mount Franklin bottle, a H2Go bottle, and a Mizone bottle - were on the market in August 2007. 62 I think the evidence establishes that, apart from the Contour Bottle, there were at least four, perhaps six, bottles with waists of varying degrees on the Australian market in August 2007. It follows that the fact that the Contour Bottle had a waist was not a unique feature in August 2007. In addition, as the respondents submitted, I think that the fact that the additional bottles have waists illustrates the legitimate desire of manufacturers and traders to use that feature otherwise than as a trade mark in the same way as widespread use of a descriptive word might indicate that it is not being used as a trade mark. 63 Ms Van Ruiten was also instructed to photograph the manner in which single serve non-alcoholic carbonated beverages were displayed for sale in refrigerators within retail outlets such as convenience stores or food service outlets. She carried out her instructions in May 2013 and 26 photographs were tendered in evidence. In a number of cases, the beverages are kept in enclosed refrigerators with glass doors or in open coolers. In other cases, the beverages are displayed on open shelves. 64 Ms Peterson is employed by CCSP as the Business Unit Counsel. She is responsible for overseeing and managing legal issues which arise as a result of the marketing, advertising, and promotion of products by CCSP under the TCCC trade marks in the South Pacific region. CCSP monitors the market in Australia for potential breaches of TCCC's trade mark rights. Employees of both CCSP and Coca-Cola Amatil (Aust) Pty Ltd are given a certain level of training and instruction about TCCC's trade marks. In November 2009, it was brought to Ms Peterson's attention that a Pepsi RAW Bottle might have been displayed for sale in Melbourne. The Pepsi RAW Bottle was a bottle sold in England which was the equivalent of the Carolina Bottle. In January and February 2010, Ms Peterson became aware that the Carolina Bottle was being offered for sale and sold in Australia. TCCC commenced this proceeding on 14 October 2010.

THE RESPONDENTS and The Carolina Bottle 65 In a number of respects, the early history of the Pepsi beverage is similar to the early history of the Coca-Cola beverage. 66 In the 1890s, Mr C Bradham, a pharmacist in New Bern in North Carolina, invented the recipe for Pepsi-Cola. The Pepsi-Cola beverage was sold under the trade mark PEPSI-COLA from 1898 and under the trade mark PEPSI from 1911. The company's trade marks now also include the word marks PEPSI LIGHT, PEPSI MAX, PEPSI TWIST, DIET PEPSI, and PEPSI NEXT, and a variety of device or brand marks. 67 In 1931, the predecessor to PepsiCo was formed under the name Pepsi-Cola Company and, in 1965, the company name was changed to its present name. 68 From 1958, PepsiCo used a glass bottle known as the "Swirl" glass bottle. Photographs of the Swirl glass bottle are shown below (at [82]). By December 1968, PepsiCo had obtained trade mark registration in the United States in relation to the shape of the Swirl glass bottle. By May 2004, PepsiCo had obtained trade mark registration in the United States in relation to the shape of the Swirl PET bottle. Mr So agreed in his evidence that PepsiCo believed, and continues to believe, that the Swirl glass bottle design and the Swirl PET bottle design are distinctive of it and its products. 69 In 1993, Pepsi Max was released as a low-kilojoule, sugar-free cola alternative to Diet Pepsi or Pepsi Light. 70 Since 2005, PepsiCo and its bottlers in at least 30 countries worldwide have launched and sold Pepsi soft drinks in glass bottles which are the same or closely similar to, in terms of shape, the Carolina Bottle which is the bottle sold in Australia. 71 PepsiCo is a worldwide distributor of soft drinks and snack products. The company, and its wholly-owned subsidiaries and affiliates, is one of the two largest soft drink companies in the world. The other company is TCCC. PepsiCo is the largest seller of salted snack products in the world. The PEPSI mark has been used consistently over the years. By contrast, the Pepsi logo has changed over the years. 72 The word marks PEPSI and PEPSI MAX have been registered as trade marks in a number of jurisdictions throughout the world and they have an international reputation. PepsiCo engages in intensive advertising campaigns in Australia for its Pepsi beverages. 73 Pepsi beverages in the Carolina Bottle are sold through similar channels as Coca-Cola beverages. Mr Van Dijk identified substantial sales through the grocery channel and the petrol and convenience channel. He also referred to substantial sales through the HORECA channel. He said that Pepsi beverages are also sold through the on-premise and licensed on-premise channel, the impulse channel, and the quick service restaurants channel. 74 Pepsi beverages are sold in PET bottles and in cans of various sizes, as well as being sold in the 300 ml glass bottle. They are sold as a single item or as multi-packs. 75 Schweppes supplies some of its trade customers with a refrigerator for the purpose of stocking Schweppes and Pepsi products. These refrigerators often have Schweppes or Pepsi branding on them. 76 Sales of the Pepsi beverages in the Carolina Bottle have been very low compared with sales of the Pepsi beverages in PET bottles or cans. There has been some trade advertising of the Carolina Bottle, but no large-scale advertising campaigns directed to consumers. As might be anticipated, the respondents' marketing is, like TCCC's, very sophisticated. 77 It is necessary to address the design process for the Carolina Bottle. This issue is relevant because TCCC submits that, in designing the Carolina Bottle, PepsiCo and PepsiCo Australia had an intention to deceive and that, insofar as this is a borderline case with respect to any of TCCC's claims, an intention to deceive tips the scales in its favour. The precise submission made by TCCC is that the respondents should be taken to have been aware of the risk of confusion by reason of its Carolina Bottle and decided to take that risk. I discuss the relevant legal principles later in these reasons. 78 The Carolina Bottle was designed in the United States for world-wide purposes and Mr Le Bras-Brown played an important role in that process. The Carolina Bottle was first introduced into the Australian market in August 2007. Mr Cumming played a part in the process whereby that occurred. When I refer to the Carolina Bottle in this context, I ignore (as did the parties) minor differences which might be due to an individual bottler's constraints or needs. 79 Mr Le Bras-Brown joined PepsiCo in June 2003. Before June 2003, PepsiCo commenced a project in the United States which was known within the company as the "Carolina project". The project involved the formulation of a design of its PET bottles to a design called the "Carolina PET". To assist in that process, PepsiCo engaged a New York design firm known as 4sight inc ("4sight"). The actual design process which resulted in the Carolina PET had been completed before June 2003. The first PET bottle with the Carolina design was a 2 litre bottle, but there followed a range of both smaller and larger sized PET bottles. An image of the Carolina PET Bottle appears below. 80 The principal design features of the Carolina PET were, first, a series of undulating horizontal wavy lines which appear above and below the label and, secondly, a significant "drawing in" of the bottle in its middle. The horizontal wavy lines were considered to be a feature which assisted in ensuring the bottle retained its shape and resisted a process called creep. Creep is a condition where a bottle takes on a rounded shape because of internal pressure. In addition, the horizontal wavy lines were considered to convey the image of Pepsi expanding by the use of sinusoidal pattern where the waves expand logarithmically. The "drawn in" middle of the bottle meant that the bottle was easier to grip and, therefore, it was easier to pour the contents. 81 The next stage of what was now known within PepsiCo as the Carolina project was the design and development of two glass bottles, one a returnable glass bottle ("RGB") and the other a non-returnable glass bottle ("NRGB"), to complement the Carolina PET. Mr Le Bras-Brown was directed to lead this stage. His design brief was to produce a design for a bottle which was an "ownable" and a proprietary design for PepsiCo. The bottle was to have an exciting look and feel that matched or complemented the existing Carolina PET bottle. Mr Le Bras-Brown said that he wanted to take the undulating horizontal wave feature and translate it to glass. He wanted to design a bottle that was uniquely Pepsi and was an iconic bottle that the public would identify as "the Pepsi bottle". 82 Mr Le Bras-Brown's understanding was that, from at least the 1980s, PepsiCo's glass bottle was called the "Swirl Bottle". A key design feature of the Swirl Bottle was twisted diagonal fluting which was described by Mr Le Bras-Brown as like a Corinthian pillar that had been twisted in the middle. He said that the Swirl Bottle had a shoulder that led into a drawn in mid-section and then a pronounced heel at the bottom of the bottle. Although the images are not particularly good, I set out below a picture (left) showing the PepsiCo Swirl glass bottle on the left of other bottles and cups. Another image of the Swirl Bottle is shown on the right. 83 Mr Le Bras-Brown said that the design of a glass bottle which is to carry a carbonated soft drink is subject to constraints, and that those constraints differ according to whether the glass bottle is to be produced as a RGB or as a NRGB. A RGB is made of relatively thick glass and the bottles are collected by PepsiCo, or one of its subsidiaries, and are reused after a process of cleaning. A NRGB is made of relatively thin glass and is not collected for reuse. 84 Both types of glass bottle must be capable of being used on a bottle production line. A RGB must be capable of being collected after use, stored, and refilled. This means there are greater design constraints in the case of a RGB. The design of a RGB must accommodate an existing bottle float (i.e., the collection of bottles being used and reused), which may be very valuable and which may include up to six prior versions of the company's glass bottle. To accommodate the existing bottle float, the designer of a bottle must have regard to the height, fill height, and diameter of existing bottles, and the likely "touch" points on the production line between the existing bottles and any proposed new bottle. For example, it is important that the touch points remain the same so that the old and new bottles can continue to be processed, washed, bottled, and stored together in the bottling plant. It is important not to place any decorative features at the touch points because they can be damaged. A bottle must also have a sufficiently low centre of gravity so that it is not easily knocked over. 85 Mr Le Bras-Brown said that there is more freedom for designers in the case of the NRGB, which are only used once. Nevertheless, there are still some constraints and they result from the dimensions of the bottling line and the size of crates and coolers. 86 The firm 4sight assisted Mr Le Bras-Brown and his team in the design of the Carolina Bottle, both the RGB and the NRGB. In broad terms, the process was that 4sight provided concept drawings to Mr Le Bras-Brown and his team, who then ruled out those concepts which they considered inappropriate and with respect to the balance, sought the comments of PepsiCo's marketing department. Drawings of preferred designs were then forwarded to 4sight, who were required to ensure that the drawings were volumetrically accurate. Once a design was selected, it was to be sent to a glass supplier in a pilot market for its views and the preparation of final engineering drawings. 87 In October 2003, the preferred designs for the Carolina RGB were those designated by 4sight as "1" (300ml RB Bottle Design Exploration: 09.24.2003) and "1C" (300ml RB Bottle Design Exploration: 09.24.2003). Images of those designs are as follows: It can be seen from these images that both of these designs have a rounded shoulder and that Design 1 has a pronounced bulge above the waist of the bottle. 88 As it happened, the final choice came down to a choice between those designs designated by 4sight as "1B" (300ml RB Bottle Design Exploration: 09.23.2003) and "IE" (355ml RB Bottle Design Refinements: 4.29.04). Images of those designs are as follows: 89 The principal difference between Design 1B and Design 1E is the shape of the shoulder of the bottle, with Design 1E being more conical and Design 1B being more similar to the silhouette of the Swirl Bottle. 90 Mr Le Bras-Brown said that he considered that one advantage of Design 1E was that it had a straight tapered shoulder and that facilitated tactiphane labelling. Tactiphane is a form of clear paper which is made using cellophane cells. The advantage of tactiphane labelling is that it gives a no label look, but in fact it can be washed off the bottle when it is returned to the bottlers. This is an advantage for bottlers because it makes the bottle generic in a labelling sense. Mr Le Bras-Brown was very keen that there should be a shoulder on the Carolina RGB that could be used for ACL labelling and also tactiphane technology, where available. 91 Mr Le Bras-Brown said that he and his team chose Design 1E over Design 1B for both aesthetic and practical reasons. They considered Design 1E to be better looking, more attractive and contemporary than Design 1B, and, with the tapered shoulder, it provided maximum flexibility with labelling options. The final decision was made by a group of people within PepsiCo, some from the design section and some from the marketing section. 92 Mr Le Bras-Brown agreed in cross-examination that he had not provided any explanation as to why the "pronounced bulge design" (shown in Design 1 and similar images with slight variations) had been rejected. 93 Mr Le Bras-Brown said that he was aware of the Contour Bottle and that he "designed away" from that bottle because he wanted PepsiCo to have its own distinctive bottle and because he did not want to become embroiled in disputes with TCCC. 94 As far as the NRGB was concerned, the design ultimately selected was similar to the Carolina PET, but that design was only appropriate for jurisdictions where the "belly band" paper label was viable. That did not include Australia and, in this country, the design for the NRGB was the same as the design for the RGB. An image of the NRGB for countries where the belt band paper label was viable is as follows: 95 Mr Le Bras-Brown said that the above design for the NRGB achieved the objective of complementing the Carolina PET design, but the RGB was not able to take those features. His evidence was that, nevertheless, both types of bottle had the "Carolina family look dominated by the waves that were in the bottle". 96 Mr Le Bras-Brown agreed that the Carolina RGB is not a generic bottle and that it is distinctive and was intended by PepsiCo to be distinctive. 97 PepsiCo made design applications throughout the world in relation to Design 1E and there are two Australian registered designs for that design. 98 There were features of Mr Le Bras-Brown's evidence which caused me to scrutinise it carefully. First, he gave a good deal of evidence about design constraints and I have summarised that evidence. However, for the most part, that evidence was not linked to particular design choices made by Mr Le Bras-Brown. Secondly, he accepted in cross-examination that he had not given any evidence explaining when and why the "pronounced bulge" design was rejected. It might, however, be inferred from some earlier evidence Mr Le Bras-Brown gave that the "pronounced bulge" design was rejected because it was too "top heavy". Thirdly, to my mind, it is undoubtedly the case that the Carolina Bottle is, in overall appearance, closer to the Contour Bottle than either the Swirl glass bottle or the NRGB where the belly band label is used. Despite these matters, I do not think that Mr Le Bras-Brown had an intention to deceive or an intention to copy the Contour Bottle. All of the designs he considered had a contoured shape. All of the design images show the bottles clearly marked with the word "Pepsi" and the globe device, and Mr Le Bras-Brown would have been aware that this is how they were displayed for sale. All of the design images also clearly show the horizontal waves and this was an important design feature and it was a link with the Carolina PET. I accept Mr Le Bras-Brown's evidence that labelling flexibility was the reason he chose the Design 1E over Design 1B. Mr Le Bras-Brown was asked whether other labelling on the NRGB designed for the belly band labelling was feasible. He said ACL labelling was not possible and that tactiphane labelling would be very expensive. ACL labelling refers to applied ceramic labelling, where a permanent ink label is fired onto glass using silk screening. Mr Le Bras-Brown's evidence about the difficulty with ACL labelling is supported by a contemporaneous email he wrote to other employees of PepsiCo on 16 August 2005. In cross-examination, Mr Hunter said that both forms of labelling would be possible, although he did not seem to have a deep knowledge of tactiphane labelling. I am not satisfied that I should reject Mr Le Bras-Brown on the basis of the very brief and, to some extent, undeveloped evidence of Mr Hunter in cross-examination. Whether Mr Le Bras-Brown was, without realising it, affected in some way by the attractive features of the Contour Bottle, I cannot say. I can say that I am not satisfied that he had an intention to deceive or cause confusion, or an intention to copy the Contour Bottle. 99 I turn now to the introduction of the Carolina Bottle into the Australian market. Prior to the introduction of the Carolina Bottle, the Pepsi beverages were sold in the Swirl bottle and that was a NRGB. 100 Until 2006, the Pepsi beverages were also sold in a PET Swirl or "Global Swirl" bottle. Sometime in 2006, PepsiCo Australia changed from the Global Swirl PET Bottle to the Carolina PET Bottle. 101 In the same year, PepsiCo Australia and Schweppes began considering the introduction of the Carolina Bottle into Australia. That was prompted by what was happening internationally and by the fact that, at that time, the moulds for the existing glass bottle design were wearing out and would soon need replacing. 102 On 24 August 2006, Mr Cumming wrote to Mr Le Bras-Brown asking for drawings and examples of the different options for the Carolina Bottle, non-returnable, and with a capacity of 300/390 ml. Mr Cumming advised Mr Le Bras-Brown that he was considering adhesive labelling or ACL printed artwork. Mr Le Bras-Brown sent Mr Cumming some drawings the next day. Those drawings showed two options, one appropriate for a paper label around the middle of the bottle, and the other was the design for the Carolina Bottle. Mr Cumming favoured the design for the Carolina Bottle because, unlike the first option, it did not involve the extra costs associated with a special label, and because he considered that it had a more conventional shape. He considered that it looked similar to the Schweppes 300 ml glass bottle. That was of some significance because the 300 ml Pepsi and Pepsi Max beverages would ordinarily be offered for sale as part of the Schweppes 300 ml glass range. 103 Ultimately, it was the marketing department of PepsiCo Australia, in consultation with Schweppes, that made the decision to adopt the Carolina Bottle. 104 As I have said, I accept Mr Cumming's evidence and I do not think that there was any intention to deceive or cause confusion or copy the Contour Bottle on his part. Nor could it be said (as TCCC put in its written submissions) that he knowingly took a risk that consumers would be confused by the Carolina Bottle. 105 In August 2007, the Carolina Bottle for Pepsi beverages was launched in Australia. In 2008, PepsiCo Australia and Schweppes experienced a problem with the glass manufacturer and they decided to change to a new manufacturer. That had two consequences. First, it led to some very minor modifications to the design of the bottle. Secondly, it resulted in the respondents selling the Pepsi beverages in Australia in a generic bottle, rather than the Carolina Bottle, between May 2008 and February 2009. They resumed selling the Pepsi beverages in the Carolina Bottle after February 2009. 106 PepsiCo Australia and Schweppes changed manufacturers again in late 2013 and again there were some very minor changes to the design of the Carolina Bottle. 107 PepsiCo has not applied to register the Carolina Bottle shape as a trade mark in the United States or any other jurisdiction. 108 In Germany, PepsiCo released a 200 ml Carolina Bottle in April 2009 and a 330 ml Carolina Bottle in May 2010. In July 2010, TCCC sent PepsiCo's German subsidiary a cease and desist letter. TCCC launched proceedings alleging infringement of certain of TCCC's registered Community and German trade marks, but was unsuccessful at first instance. TCCC has appealed, but the appeal has not yet been heard. 109 In New Zealand, PepsiCo released beverages in the Carolina Bottle in 2009. TCCC commenced proceedings against PepsiCo and Frucor Soft Drinks Ltd alleging trade mark infringement, passing off, and breaches of the Fair Trading Act 1986 (NZ) in relation to the sale of Pepsi beverages. Again it was unsuccessful at first instance (Coca-Cola Co v Frucor Soft Drinks Ltd and Another (2013) 104 IPR 432; [2013] NZHC 3282). TCCC had lodged an appeal. 110 TCCC has not complained about PepsiCo's use of the Carolina Bottle in any other jurisdiction. 111 Ms Evans' duties include supervising and overseeing the operation of the Consumer Information Centre at PepsiCo Australia and reporting details of complaints and feedback to the management of the company. She has access to PepsiCo Australia's books and records in relation to complaints and feedback received in the ordinary course of business. Her evidence establishes, and I find, that PepsiCo Australia has no record of having received any complaints or feedback regarding the shape of the Carolina Bottle, or any similarities between it and any other product. 112 Ms Vo's evidence establishes that, as exclusive bottler of PepsiCo's products in Australia, Schweppes also handles any complaints or feedback regarding those products, and that Schweppes has a fairly sophisticated system for recording complaints and feedback from consumers and for bringing those complaints or that feedback to the attention of the appropriate person within Schweppes. Her evidence establishes that no complaint about the shape of the Carolina Bottle, or any similarities or confusion between it and any other product, has been recorded by Schweppes.

Matters of Design 113 Mr Crone is an industrial designer who has practised in the field of industrial design for over 40 years. He is the Managing Director of UNO Australia Pty Ltd, which is a design consultancy firm based in Melbourne. The firm provides design services in branding, packaging and visual communication, product design and engineering, and interior design for "retail and exhibitions". 114 Mr Crone outlined the process a designer would follow in the ordinary case in designing packaging for a client. It is not necessary for me to set out the details. One point which Mr Crone made in the course of giving his evidence, and which is contentious, is that consumers recognise colour and shape before they recognise symbols and words. 115 Mr Crone identified various features of the Contour Bottle and expressed the opinion that the most significant aspect of the appearance and packaging of the Contour Bottle was the low waisted contour shape as seen in profile, that is, the silhouette of the bottle. He expressed the opinion that the overall shape of the Contour Bottle is a major contributor to the easy recognition of the product. In considering a bottle where ACL labelling is used and the words appear in the Spencerian script, Mr Crone said that the dominant colour was the dark brown colour of the contents of the bottle. Mr Crone said that the labelling (i.e., the Spencerian script) and the decoration flutes were also important design features of the Contour Bottle, but would be noticed after the overall low waisted contour shape and dark brown colour. The type and colour of the cap or top was also an important design feature. 116 Mr Crone expressed the opinion that, from a design perspective, the signs constituting TCCC's four trade marks reflect as their single most recognisable element the overall low waisted contoured shape with its proportions. 117 Mr Crone was provided with the third party bottles purchased by Ms Van Ruiten. He examined these bottles and concluded that the Contour Bottle was of a significantly different appearance and that those third party bottles were all substantially different for one reason or another. He said that only three of the third party bottles - the Lucozade Bottle, the Schweppes Bottle, and the V PET Bottle - had a waist of some sort. However, for reasons he gave in evidence, he expressed the view that that is where the degree of resemblance with the Contour Bottle ends. 118 In cross-examination, Mr Crone agreed that the Lucozade bottle had a "lowish" waist, "somewhat", and a "somewhat" contoured shape. He agreed that the Schweppes Bottle and the V PET Bottle had waists and, in the case of the former, a minor contour. He was shown the additional bottles and he agreed that a number of them had waists. Mr Crone acknowledged that, whilst some of the third party bottles were similar in colour to the Contour Bottle due to the colour of the glass or the contents of the bottle, they differ significantly in other respects due to the colour of the labelling or the shape of the bottles. 119 Mr Crone expressed the opinion that the most significant element of the appearance of the Carolina Bottle from a design perspective is the overall contour shape highlighted by the low waisted shape with its thinnest diameter located in the lower section of the bottle as seen in profile. Other significant design features are the ACL labelling and the surface decoration with a wave design, although he considers that "these form part of the slower rational response from the consumer". 120 Mr Crone said that the colour of the caps used for the Carolina Bottle - blue for regular and black for the sugar-free variety - were also a significant feature. 121 Mr Crone identified the following similarities between the Contour Bottle and the Carolina Bottle: the overall silhouette profile (which included the low waisted contour shape), the ACL labelling, and the dark brown colour of the bottles, that being a result of their contents. He also noted that the same cap colour is used for the sugar-free varieties. He identified the following differences between the two bottles: the Contour Bottle has a slightly curved neck area, the Coca-Cola logo, and vertical surface decoration, whereas the Carolina Bottle has an almost straight sided top neck area, the Pepsi logo, and the horizontal wave decoration. The colour of the caps was different in the case of the regular beverage. He considered that the similarities were significant from a design perspective because, in his opinion, colour and shape are the features of design that will be recognised by consumers upon first seeing the products in the market place. He expressed the opinion that the overall impression is similar because the "silhouette shape is broadly the same with what I regard as only a slight difference in the top of the bottle". 122 As far as TCCC's four trade marks were concerned, after identifying certain differences, Mr Crone expressed the opinion that an overall impression of similarity based on the shape of the silhouette persisted and that the similarity was stronger in the case of TM 1160893 and TM 1160894 because of the absence of surface decorations in the case of those trade marks. Mr Crone produced a number of overlays which enable one to make a visual comparison between the outline or silhouette of the Carolina Bottle and the Contour Bottle and each of the TCCC's four trade marks. 123 Mr Crone was also asked by TCCC to review a file which had been provided to him and which was described by those instructing him as "Pepsi Design File". He drew two conclusions from his review of that file. First, he said that the design directions at the outset of the project, which were to design a bottle which would distinguish the Pepsi brand from its competitors, including TCCC, were not followed. Secondly, there were no appearance attributes of the Carolina Bottle which constituted a family resemblance with the Carolina PET bottles apart from the surface decoration (i.e., the horizontal waves). 124 Mr Crone also expressed the opinions that a family resemblance between the Carolina Bottle and the Carolina PET Bottle could have been achieved, and that the Carolina Bottle did not draw on the heritage of the earlier Pepsi Swirl Bottles. 125 Mr Hunter is a design consultant who has qualifications in mechanical engineering. He has extensive experience in packaging designs, including the design of bottles and other containers for beverages and other liquids. 126 Mr Hunter disagreed with Mr Crone about the importance of colour and shape as a visual cue for product differentiation. He considered that label graphics and, in particular, branded logos, are the most important features in terms of enabling a consumer to differentiate and locate a package on a shelf. Furthermore, Mr Hunter did not agree with Mr Crone that surface texture was necessarily a less important visual cue than colour and shape. 127 Mr Hunter identified a number of design constraints in the case of a glass container for a carbonated beverage. First, he said that the container must necessarily be of a round shape and that, at what he called the geometry transition points (i.e., shape changes, such as those from the body of the bottle to the shoulders, and the shoulders to the neck of the bottle), gentle radiusing or curvature of the shape must be applied to avoid stress concentration. Secondly, he said that merchandising or marketing requirements may affect the shape of a bottle. For example, a bottle will have to fit into existing refrigeration units, or, for marketing purposes, the bottle will need to be of a certain volume. Thirdly, he said that carbonated beverage containers must be designed to be filled at very high speeds, which means, ordinarily at least, that the diameter of the bottle at the base of the shoulder must be the same as the diameter of the bottle near its base in order to prevent the bottle from tipping over. Fourthly, he said that there are design limitations associated with placing a label or texture on the bottle. Fifthly, he said that there are design constraints resulting from the type of closure (i.e., a crown seal or a screw top) to be used. Finally, Mr Hunter said that ergonomics, such as the ability to hold the bottle while consuming its contents, are relevant to design. 128 Mr Hunter said that he did not agree with Mr Crone that there is anything distinctive about the dark brown contents of the Contour Bottle or the labelling information on the back of the Contour Bottle because those features are common to other bottles. 129 Mr Hunter did not agree with Mr Crone that the low waist of the Contour Bottle is the most significant feature of its appearance. He thinks that it is distinctive in its appearance as the result of a unique combination of a number of features which it is not possible to separate out. Those features are the low waist in the body area (i.e., the minimum of the pinched waist is in the lower third of the body), the serpentine shaped neck (i.e., a curved transition from a concave shape to a convex shape), the vertical fluting in the body and neck of the bottle, the recessed convex curved label area in the shoulder of the bottle which is not fluted, and the usage of the distinctive Coca-Cola cursive script logo in the label area. Mr Hunter said that the Spencerian script logo is one of the world's most recognisable visual symbols. 130 Mr Hunter did not agree with the suggestion that the provision of a waist "per se" was a unique feature of the Contour Bottle and he expressed the opinion that the provision of a waist to provide an ergonomic benefit in terms of holding a bottle had been used in other carbonated beverage products for a long time. He referred to one of PepsiCo's early bottles, the "Peanut Bottle", which had a waist. However, as TCCC pointed out, there is no evidence that the Peanut Bottle was ever sold in Australia. Mr Hunter also said that the Pepsi Swirl Bottle also had a waist. 131 Mr Hunter expressed the opinion that most of the features previously identified as distinctive features of the Contour Bottle were features of the trade marks. There are exceptions to that opinion, in that, for example, the vertical fluting is not shown in TM 1160893 and TM 1160894. 132 Mr Hunter reviewed the third party bottles identified by Mr Crone. He expressed the opinion that none of them have the five features of the Contour Bottle (see [129]) which he says in combination produce its distinctive appearance. He noted that three of the bottles - the Lucozade Bottle, the Schweppes Bottle, and the V PET Bottle - have waists. 133 Mr Hunter expressed the opinion that the most prominent and unique design feature of the Carolina Bottle, other than the Pepsi brand and logo itself, is the horizontal wave feature, which he described as a surface texturing of raised ridges or bumps located in the mid-portion of the bottle. 134 Mr Hunter could discern only one similarity between the Contour Bottle and the Carolina Bottle, and that was that they each have waists in the body of the bottle. He said that the waist in the Carolina Bottle is not a low waist like the waist of the Contour Bottle, and he considered the waist in the Carolina Bottle less "pinched". 135 The design evidence was helpful in educating me as to the relevant design considerations in the case. However, beyond that I do not find it of great assistance in resolving the issues in this case. I prefer Mr Hunter's evidence over that of Mr Crone, largely because, once I had been educated on what to look for, Mr Hunter's evidence accords with my own views of the relevant issues (see Cat Media Pty Ltd v Optic-Healthcare Pty Ltd (2003) ATPR 41-933; [2003] FCA 133 at 46,937 per Branson J). 136 I should identify some particular opinions of each witness which I accept, and others that I do not accept. 137 With respect to Mr Crone's evidence that consumers recognise colour and shape before they recognise symbols and words, I accept that evidence as to colour, but not as to shape. In the case of cola products, I accept that the colour of the labelling or writing on the bottle or caps can be a very important visual cue. I am not satisfied that the same can be said of shape. It seems to me that it will depend on the circumstances. For example, the words "Coca-Cola" in Spencerian script may be recognised as an image well before the stage at which they can be read. 138 I accept Mr Hunter's evidence in preference to that of Mr Crone as to the significant features of the Contour Bottle and of each of TCCC's four trade marks. It accords with my own opinion on these matters. Furthermore, it accords with TCCC's Brand Identity and Design Standards (see [35] above). 139 Mr Crone's overlays were helpful in showing the similarities and differences in the outlines and silhouettes of the Contour Bottle and TCCC's four trade marks on the one hand, and the Carolina Bottle on the other. However, lining them up by reference to the top of the bottle or mark, or by reference to the bottom of the bottle or mark, seemed to produce different results and they do not give one a clear appreciation of what is seen by the naked eye, that is, that the waist of the Carolina Bottle appears to be higher up the bottle and of a more gently curving nature than the apparently lower pinched waist of the Contour Bottle. Furthermore, in view of the relevant legal principles which are discussed below, their use is limited. 140 I accept Mr Hunter's evidence in preference to that of Mr Crone as to the weight to be accorded to the similarities and differences between on the one hand, the Contour bottle and TCCC's four trade marks and, on the other, the Carolina Bottle. 141 With respect to Mr Crone's evidence as to the design process for the Carolina Bottle, I have already set out my findings in relation to that matter based on the direct evidence of Mr Le Bras-Brown and Mr Cumming. 142 I have a reservation about one aspect of Mr Hunter's evidence and that relates to his emphasis on the waist of the Pepsi Swirl glass bottle. I think that he overemphasised the waist on that bottle and that it is really a much less significant waist than the waist of the Carolina Bottle. I do not think his evidence about the distinctive features of TM 63697 differs in any material respect from the features identified by the Full Court in Coca-Cola Company v All-Fect Distributors Ltd (1999) 96 FCR 107 ("Coca-Cola v All-Fect Distributors") (see [233] below).

Matters of Consumer Behaviour 143 Dr Gibbs is a behavioural scientist and his main report is an annexure to his affidavit sworn on 16 July 2013. In his second affidavit sworn on 13 January 2014, Dr Gibbs responds to the opinions expressed by Professor Klein. 144 Dr Gibbs' expertise is in the field of behavioural science and marketing. He has a Bachelor of Science degree in biopsychology from the University of British Colombia, a Master's degree in cognitive psychology from the same university, and a PhD in behavioural science and marketing from the Graduate School of Business at the University of Chicago. Since the early 1990s, he has conducted an independent research-based consulting practice, and he states that, since 2009, he has conducted all of his work, including his academic research and university teaching, under the "auspices of his consultancy, Behavioural Science Insights". Dr Gibbs has been a faculty member in the Melbourne Business School at the University of Melbourne since 2003 where he is Principal Fellow in Marketing and Behavioural Science and he holds the title of Associate Professor. Further details of his qualifications and experience are set out in his report. 145 Professor Klein is Professor of Marketing at the Melbourne Business School at the Melbourne University, and she has held that position since 2009. From 2012 to 2013, she was also Associate Dean of the full-time Master of Business at the Melbourne Business School. Professor Klein has a Degree of Bachelor Arts, Majoring in Psychology, from the University of South Florida, a Degree of Master of Science, Majoring in Counselling Psychology, from the State University of New York-Albany, and a PhD in Social Psychology from the University of Michigan. In her current position, Professor Klein gives lectures in decision-making and marketing management, and she has previously given lectures in consumer behaviour, advertising and marketing communications, and marketing research. Further details of her qualifications and experience are set out in her affidavit. 146 The respondents submitted that I should reject the tender of Dr Gibbs' main report. They did not submit that Dr Gibbs' evidence did not constitute a relevant field of expertise, or that Dr Gibbs lacked the expertise to express the opinions he set out in his report and affidavit (ss 76(1) and 79 Evidence Act 1995 (Cth); Sydneywide Distributors Pty Ltd v Red Bull Australia Pty Ltd (2002) 55 IPR 354; [2002] FCAFC 157 ("Sydneywide Distributors v Red Bull Australia"); Cadbury Schweppes Pty Ltd v Darrell Lea Chocolate Shops Pty Ltd (2007) 159 FCR 397). Rather, their principal contention was that Dr Gibbs' report was of no assistance to the Court because his starting point was the same as his finishing point. They relied on two related contentions in support of their principal contention. 147 I overruled the respondents' objection to Dr Gibbs' report and it was received into evidence. My reasons for that decision follow. 148 The respondents submitted that Dr Gibbs started his analysis by assuming, or acting on the basis, that TCCC's Contour Bottle and the respondents' Carolina Bottle were similar because both had a contoured shape. They argued that it was implicit in this starting point that consumers were likely to be misled by the similarity in shape between the two bottles. The respondents submitted that that was in fact Dr Gibbs' conclusion, that is to say, that consumers were likely to be misled by the similarity in shape between the two bottles. 149 The respondents pointed to the title of Dr Gibbs' report, "Marketplace Consequences of the Packaging of Pepsi in a Contour Bottle like Coke's: The Perspective from Behavioural Science and Marketing", in support of their contention. They also pointed to the description of Section G of Dr Gibbs' report, "Analytical Framework: Behavioural Consequences of Attribute Overlap Across Brands", and the fact that, in the first paragraph of that section, Dr Gibbs stated that his premise was that Pepsi has adopted a bottle similar to TCCC's Contour Bottle. 150 The two related contentions upon which the respondents relied were as follows. First, they submitted that, if Dr Gibbs held the view that the two bottles were similar because of their contoured shape, that view was no more than Dr Gibbs' lay opinion and it was not based on any expertise he had or any empirical evidence made available to him. The respondents submitted that the field work Dr Gibbs did in relation to two of the questions he was asked to consider was very limited and of no consequence. Secondly, they submitted that Dr Gibbs' report was extremely confusing. For example, in his affidavit in reply, he states that Section F of his report, which is entitled "Some Key Foundational Concepts", and Section G, to which I have already referred, did not themselves express his expert opinions on the issues in the proceeding, or should not be treated as if they were conclusions about specific issues in the case. The respondents submitted that, despite that statement, there were paragraphs in those sections in which Dr Gibbs does express opinions on specific issues in the case. They identified those paragraphs in the course of submissions. 151 I decided that the matters which the respondents raised went to the weight to be given to Dr Gibbs' report, and not to its admissibility. Although it is not until well into his report, Dr Gibbs does set out the reasons for his view that the two bottles are similar because of their contoured shape (paragraph 78). Although he relies, in part, on his own view as to the similarities in the appearance of the two bottles, he also purports to rely on expertise he claims he has in the field of sensory psychology. It is true that his opinions are not based on empirical evidence by way of surveys or tests, but that fact is not fatal to admissibility (Australian Postal Corporation v Digital Post Australia Ltd (2013) 105 IPR 1; [2013] FCAFC 153 ("Australian Postal Corporation") at 9, [28]). It is also true that there is a level of confusion associated with Sections F and G of Dr Gibbs' report (and other sections for that matter), but those sections are not incomprehensible, and the matters the respondents identified were relevant to weight, and not admissibility. 152 I turn now to summarise briefly Dr Gibbs' report and his affidavit in response to Professor Klein's affidavit. 153 Dr Gibbs' main conclusion is that consumers making purchasing decisions in the context of fast moving consumer goods (which these cola products are) are likely to be misled by the shape of the respondents' Carolina Bottle. This means misled by the shape of the Carolina Bottle in relation to the Contour Bottle. Dr Gibbs said that a consumer is likely to make one or more of four information processing errors, which he identifies as "misidentification", "miscuing", "misinference", and "misassociation". These "errors" form part of a construct developed by Dr Gibbs, with misidentification and miscuing bearing primarily on choice, rather than judgment, and the other two (i.e., misinference and misassociation) bearing primarily on judgment, rather than choice. Furthermore, according to Dr Gibbs' construct, misidentification and misinference are primarily conscious, whereas miscuing and misassociation are primarily nonconscious. 154 Dr Gibbs expresses the opinion that consumers are likely to be misled by incorrectly identifying the Carolina Bottle as the Contour Bottle "due to the mutual use of the contoured bottle by the two brands". Dr Gibbs said that by contoured shape he meant an "out in out", or "hour glass", shape. He said that the presence of brand names and logos is unlikely to act as a correcting factor. He also said that, in his opinion, miscuing - a primarily nonconscious process where the consumer takes the contour shape as a behavioural cue - is likely. 155 Dr Gibbs also said that, in his opinion, misinference is likely to occur. As I understand Dr Gibbs' construct, this is where the consumer knows the identity of the brand he or she is purchasing, but draws an incorrect inference from what Dr Gibbs referred to as the mutual use of the Contoured Bottle. The only example of an error constituting "misinference" given by Dr Gibbs was that consumers might infer that PepsiCo and PepsiCo Australia were subsidiaries of TCCC. I find that suggestion fanciful in the extreme in view of the well-known rivalry between the two companies. 156 Dr Gibbs also said that, in his opinion, the mutual use of the Contoured Bottle is likely to lead to misassociation, which is the primarily nonconscious linking by the consumer of TCCC's brand associations with the respondents. 157 One of Dr Gibbs' main conclusions is that the Contour Bottle and the Carolina Bottle are perceptually similar, and that "most consumers, under most circumstances, would treat these bottles as similar or even as the same in the course of their marketplace information processing". Other than relying on his personal subjective response, in respect of which he correctly recognised there were limitations, Dr Gibbs relied on a level of expertise in sensory psychology and experience using low-level sensory stimuli. He said that he "probably" has more expertise in the area than the "average consumer-behaviour academic", although his expertise in "this specific area" is not at the same level as his expertise in the other areas in his report. 158 Dr Gibbs' report is directed to TCCC's claims of passing off and contraventions of the Trade Practices Act. He does not expressly address the issues under the Trade Marks Act. This is an observation, not a criticism (see the remarks of the Full Court in Australian Postal Corporation at 10, [29]-[30]). 159 I turn now to summarise briefly the evidence of Professor Klein. 160 Professor Klein identified the matters in Dr Gibbs' report with which she agreed and those with which she disagreed. She comments on Dr Gibbs' opinion to the effect that the shape of the Contour Bottle is atypical and that it would be used as a brand cue, particularly in the case of cola products which are convenience goods where consumer involvement with the purchase is low. Dr Gibbs described low consumer involvement in a purchase as those cases where consumers acquire information narrowly and rely more heavily on brands and "salient brand cues" to expedite product selection, rather than on more detailed information that may be present, "and buying behaviours can become habitualised or automated responses to these cues". Professor Klein responds to these opinions by making the following points. First, she states that a great deal of research has been done on how consumers categorise various stimuli. Secondly, she states that a stimuli which is a cue is known as a diagnostic cue where it has, what she called, high cue validity. Thirdly, she expresses the opinion that a brand name, such as Coke or Pepsi, in the case of those cola beverages, would be a highly diagnostic cue. Colour might also be a highly diagnostic cue. Fourthly, she expresses the opinion that the most diagnostic cues will be relied on by the consumer and, in the case of Coke and Pepsi products, that will be the brand name, logo, and brand colour. 161 Professor Klein notes that Dr Gibbs' assertion that consumers would consider the two bottles to be similar is not supported by any empirical evidence. Furthermore, she notes that the term "similar" is imprecise. Professor Klein criticises Dr Gibbs' construct of misidentification, miscuing, misinference, and misassociation. In her opinion, Dr Gibbs does not put forward any empirical evidence in support of his assertions and, in any event, the most likely diagnostic cues are the brand name, logo, and brand colour, rather than the contour shape of the bottle. Professor Klein then identifies a number of statements in Dr Gibbs' report which, in her opinion, could be established by empirical evidence and which she considers should not be accepted in the absence of observations or experiments or both. 162 A passage in Professor Klein's evidence which captures her response to a number of Dr Gibbs' opinions is as follows: In this specific setting, which cue is most diagnostic is an empirical question and could be tested. Based on a great deal of research on visual perception and categorisation, one would certainly hypothesise that the brand label would be the most important and diagnostic cue used in identifying whether the product is a COKE or a PEPSI. 163 Professor Klein identified the observations and experiments that could be carried out with a view to producing information which would be useful in determining the issues in this case. First, she said there were eye tracking studies which identify where the eyes are looking when a package is viewed. By these studies, it is possible to determine the percentage of consumers who miss both the brand name and colours on a bottle, but focus on other aspects of the bottle, such as the edges or shape of the bottle. Secondly, there are recognition and categorisation tests which have been used to determine consumer confusion. Professor Klein described how these tests are carried out: The image of each bottle could be presented on a screen for a short period of time and the consumer could be asked to indicate the maker of the product. 164 Professor Klein accepted in cross-examination that a consumer who enters a convenience store intending to purchase a fast moving consumer good may have a low involvement with the goods themselves and may notice some, but not all, of the visual cues relevant to the identification of the goods. She agreed that there is a concept or process of cognitive miserliness, that is to say, a consumer in certain cases will spend no more intellectual resources than are necessary to identify the goods he or she wants. She also agreed that there is a notion of confirmation bias where a consumer tends to see what he or she expects to see. She said that the concept "works best" where there is an ambiguity. She reiterated the evidence in her affidavit that in this case the most efficient and diagnostic cue is the names on the bottles. 165 In cross-examination, Professor Klein was referred to two passages in an article which she had referred to in her affidavit: Kapferer (1995) entitled "Brand Confusion: Empirical Study of a Legal Concept". 166 The first passage is as follows: Sometimes the design of the packaging itself is a brand asset, as for the classic Coca-Cola Bottle or the round orange-like Orangina Bottle. In the context of fast-moving consumer goods, much routinization exists: many brand-loyal consumers rely on visual cues to locate their brand among others in the shelves, and pick it in a split second without checking whether they got the right brand. 167 Professor Klein said that she agreed with these statements, although she did not think Kapferer was stating that the shape of the bottle would necessarily be the most diagnostic cue, or what people use to make their choice. 168 The second passage is as follows: In fact, with fast-moving consumer goods, purchasers become rather routinized, and one can hardly speak of a decision process (Olshavasky & Grandbois, 1979). Hoyer (1984) found little evidence of extensive, in-store decision making during laundry detergent purchases. Repeat purchase for these products is a low involvement behaviour and is done on the basis of habit. The consumer looks for the recognisable signs attached to his or her regular brand (i.e. the bottle with the yellow label and a red cap). The level of attention is often limited: such recognition needs no full attention. 169 Professor Klein said that she agreed with these statements, but with the qualification that the distinction between what is a "decision process" and whether recognition constitutes a decision process is "a bigger question than what is alluded to here". 170 Kapferer's observations are clearly relevant to the context in which TCCC's products and those of the respondents are sold. 171 For the reasons which follow, I am not prepared to place any weight on Dr Gibbs' opinions except to the extent that they are supported by Professor Klein's evidence. 172 First, I found the organisation and structure of Dr Gibbs' report confusing. He seems to have formed the view that the bottles were similar in that both had a contoured shape and then addressed the degree of similarity. 173 Secondly, I think Dr Gibbs' opinion that the shape of the two bottles would be seen by consumers as very similar was no more than his personal opinion and was not based on any expertise which he has. His reference to "some" education in the area of sensory psychology and "some" research experience using low-level sensory stimuli was vague and general and quite insufficient for me to conclude that there was a scientific basis for his understanding. His general response to the criticism that his opinions were not based on empirical evidence was unconvincing. For example, in his affidavit in reply he said: In paragraph 23, and echoed elsewhere in the Klein affidavit, it is argued that expert opinion offered to the Court should be based on empirical evidence rather than on the expert's own judgment. I believe this dichotomy is unhelpful. If appropriately formed, an expert's judgments are themselves based on empirical evidence, in the form of the expert's knowledge base of research results [emphasis added], which gives rise to what I will term his or her scientific understanding [emphasis in original] of the case-relevant phenomena. 174 In Dr Gibbs' case, the "knowledge base of research results" was never identified, nor were any of the matters which would facilitate a reliable assessment by the Court of the weight to be placed on his "scientific understanding" in the particular circumstances of this case. Dr Gibbs agreed that, in terms of the similarity between the bottles, empirical evidence would be needed for a direct situation or specific measurement. 175 Thirdly, I am satisfied that there are tests, observations, and experiments which would provide empirical evidence with respect to a number of issues addressed by Dr Gibbs. In this particular case, the absence of empirical evidence means that I have a real doubt as to whether many of Dr Gibbs' opinions have any scientific basis. For example, he said that the shape of the Contoured Bottle was a relatively potent visual stimulus and that that opinion was based on his scientific understanding. However, the content of Dr Gibbs' scientific understanding was never identified and, in those circumstances, I place no weight on this opinion of Dr Gibbs. 176 I accept that there can be difficulties with various forms of empirical testing. The problems with survey evidence are well-known (see, for a recent example, Australian Postal Corporation at 10-11, [33]-[49]). I accept that there may well be limitations on the use which can be made of the results of the tests identified by Professor Klein but, at the same time, I accept that these tests are available and can be used to establish, or at least support, the proof or otherwise of a number of propositions advanced by Dr Gibbs. In his cross-examination, Dr Gibbs agreed that it would be possible to conduct a "situation specific study". 177 Fourthly, Dr Gibbs was cross-examined about a number of footnote references he gave in his report. I was left with the impression that he was not as careful as he should have been in checking the references he gave for some of the propositions in his report. I refer, for example, to his cross-examination at transcript 226 (footnote 40), 249 (footnote 47), and 260 (footnote 48). 178 Fifthly, I found some of Dr Gibbs' answers unsatisfactory or, at least, such that I could have no confidence in his opinions. For example, he expressed the opinion that misidentification in the marketplace was likely to be "significant" because of the mutual use of the Contoured Bottle. It was put to him that he could not express that opinion in the absence of "some actual numbers". He denied that. It was then put to him that he would not be able to make the assertion in a refereed paper without doing some significant empirical work. His answer was as follows: I might. Yes. It depends what the point of the - of the - the point I was making was. Even the suggestion that an opinion on an important point in this case might be expressed in Court, but might not be included in a refereed paper, is of concern to me. 179 Sixthly, I found it surprising that Dr Gibbs could not recall whether a judge of the High Court of England had described his evidence in a case before that judge as "unreal" and "wholly impenetrable". It is difficult to believe that any expert would be unable to recall whether or not there had been such a strong criticism of his reputation and expertise. 180 Seventhly, in the Cadbury Schweppes litigation, Dr Gibbs described consumer involvement in buying chocolate which is a convenience good as low to moderate, whereas he described consumer involvement in the purchase of cola products as low. The issue is not so much whether there is a difference between the two goods, but whether Dr Gibbs had a scientific basis for saying that there was a difference. I am satisfied that he did not have such a basis. 181 Finally, I think an example of how far Dr Gibbs was prepared to go without properly checking the available material was his assertion in paragraph 84 of his report that the similarities between the bottles might lead to a misidentification of the brand being advertised in television commercials, print advertisements, and on billboards. He was prepared to make that statement without asking to see any of the relevant advertising material.

The Context in which the contour bottle and the carolina bottle are sold 182 The context in which the two bottles are sold is important to both the trade mark issues and the issues involved in the passing off and Trade Practices Act claims. The context is the same in the case of each bottle. 183 The Coca-Cola beverages in the Contour Bottle and the Pepsi beverages in the Carolina Bottle are single serve beverages designed for consumption shortly after purchase. The products are relatively inexpensive products. 184 From the retailer's point of view, the Coca-Cola beverages in the Contour Bottle and the Pepsi beverages in the Carolina Bottle are fast moving consumer goods. This means that they are stored and sold over a short period of time. 185 From the consumer's point of view, the purchase of items such as soft drinks and chocolates are considered impulse purchases and the items are referred to as convenience goods. The purchase of a Coca-Cola beverage in the Contour Bottle or of a Pepsi beverage in the Carolina Bottle is likely to be a repeat purchase, that is, one that has been undertaken before, perhaps on numerous occasions. The class of consumers likely to purchase a Coca-Cola or a Pepsi beverage is very broad, with a large proportion of the community having purchased a Coca-Cola or a Pepsi beverage at some stage. The consumers in the class will vary in intelligence and in their ability to attend to the purchase. For many consumers, the purchase will be a low involvement purchase, that is to say, they will not devote a great deal of time and attention to the purchase. They may have regard to only one cue and make their selection on the basis of that cue ignoring, or failing to appreciate, in some cases a contrary cue. In order that this observation is not misunderstood, I would add that low consumer involvement does not mean that consumers, or a substantial proportion of them, may not have a clear and strong preference for one brand over another, for example, a Coca-Cola beverage over a Pepsi beverage or vice versa. 186 The fact that the Contour Bottle has a waist was not a unique feature in August 2007, and is certainly not a unique feature today. 187 As I have said, the Coca-Cola beverages in the Contour Bottle and the Pepsi beverages in the Carolina Bottle are sold through a number of different trade channels. The trade channels which were emphasised by TCCC were those sales which occur where the consumer self selects the beverage he or she wants from a refrigerator or open air cooler. The photographs taken by Ms Van Ruiten provide examples of the circumstances in which a consumer will proceed to select the beverage of his or her choice. There was evidence in this case that there is a market for carbonated beverages and that cola products are a segment of that market. I accept that evidence about the market for present purposes. There was also evidence that cola products may be grouped together in a refrigerator or cooler. Furthermore, Mr Van Dijk said that Coca-Cola beverages are often grouped together and I accept that evidence. To a point, it is confirmed by Ms Van Ruiten's photographs. The Coca-Cola products will often include the Coca-Cola beverage, not only in the Contour Bottle (including the regular and the reduced and no sugar varieties), but also in PET bottles and cans. The Pepsi products will be displayed in a similar way. That may be seen in the planograms produced by Mr Van Dijk. A planogram is an image showing the manufacturer's preferred arrangement of beverages in a refrigerator, although there was evidence before me that retailers do not always adhere to those arrangements. I accept that, on occasions, the area surrounding the refrigerator or cooler may not be particularly well lit, and that the bottles may not be positioned so that the word mark and the logo are directly visible to the consumer. Having said that, it is also to be noted that there may be other advertising in the outlet, and even on or in the refrigerator, including colour and signs by the relevant cola manufacturer. More importantly, all of TCCC's bottles have red on them in one form or another and, when grouped together, that is a noticeable feature, even from a distance. The regular Coca-Cola beverage in the Contour Bottle has a red cap or a red belly band label and, in the case of the reduced or no sugar varieties, the Spencerian script is red.

General 188 As I have said, the goods in respect of which TCCC's trade marks are registered are non-alcoholic beverages falling within class 32. It is common ground that, if there has been a trade mark use by the respondents in offering for sale and selling the Pepsi beverages in the Carolina Bottle, then it is in relation to goods in that class. 189 For the purposes of determining questions of infringement, the relevant subsection of s 120 of the Trade Marks Act is sub-s (1), which is in the following terms: 120 When is a registered trade mark infringed? (1) A person infringes a registered trade mark if the person uses as a trade mark a sign that is substantially identical with, or deceptively similar to, the trade mark in relation to goods or services in respect of which the trade mark is registered. (Notes omitted). The key concepts in the subsection are "sign", "uses as a trade mark", and "substantially identical with, or deceptively similar to". "Sign" is defined in s 6 to include: ... the following or any combination of the following, namely, any letter, word, name, signature, numeral, device, brand, heading, label, ticket, aspect of packaging, shape, colour, sound or scent. 190 The shape of a bottle or the outline or silhouette of a bottle are capable of being signs within the definition of that word by reason of the fact that "aspect of packaging" and "shape" are included in the definition. 191 Section 17 of the Trade Marks Act contains the definition of a trade mark and it provides guidance as to the meaning of the concept in s 120(1) of use as a trade mark. The section is in the following terms: 17 What is a trade mark? A trade mark is a sign used, or intended to be used, to distinguish goods or services dealt with or provided in the course of trade by a person from goods or services so dealt with or provided by any other person. (Notes omitted). 192 TCCC does not allege that the shape of the Carolina Bottle, or its outline or silhouette, are substantially identical with any of its trade marks. Its case is that the shape of the Carolina Bottle, or its outline or silhouette, is deceptively similar to one or more of its four trade marks. Section 10 of the Trade Marks Act contains a definition of "deceptively similar", which is in the following terms: 10 Definition of deceptively similar For the purposes of this Act, a trade mark is taken to be deceptively similar to another trade mark if it so nearly resembles that other trade mark that it is likely to deceive or cause confusion. 193 Although the Carolina Bottle is a three dimensional object and the representations in TCCC's trade marks are in two dimensional form, there is no dispute between the parties that a three dimensional object may infringe a registered trade mark where the representations of the trade mark are in two dimensional form: In Re Wolanski's Registered Design (1953) 88 CLR 278; Koninklijke Philips Electronics NV and Another v Remington Products Australia Pty Ltd (1999) 91 FCR 167 ("Koninklijke Philips v Remington (1999)") at 177, [30], per Lehane J; Koninklijke Philips Electronics NV and Another v Remington Products Australia Pty Ltd (2000) 100 FCR 90 (Koninklijke Philips v Remington (2000)") per Burchett J at 105 (with whom Hill and Branson JJ agreed). 194 As I have said, TCCC's case is that the respondents have used the shape of the Carolina Bottle, or its outline or silhouette, as a trade mark, and that it is deceptively similar to one or more of its four trade marks. That case gives rise to two main issues, which are whether the respondents have used the shape of the Carolina Bottle, or its outline or silhouette, as a trade mark and, if that issue is decided in favour of TCCC, whether the shape of the Carolina Bottle, or its outline or silhouette, is deceptively similar to one or more of TCCC's four trade marks. 195 The respondents advanced, in the alternative to one another, a number of submissions in response to TCCC's case. First, they submit that the shape of the Carolina Bottle, including any aspect of the shape, is not being used by them as a trade mark. Secondly, they submit that, if, contrary to their first contention, the shape of the Carolina Bottle or any aspect of the shape is being used by them as a trade mark, it is being done in conjunction with the more traditional trade marks used by the respondents, namely the word mark and the logo. The respondents do not gainsay the proposition that two or more trade marks may be used in relation to goods: Anheuser-Busch Inc v BudejovickÝ Budvar, NÁrodnÍ Podnik and Others (2002) 56 IPR 182; [2002] FCA 390 at 228 per Allsop J (as his Honour then was). However, they submit that, if the shape of the Carolina Bottle is used by them as a trade mark, then it is as part of a combination which includes the word mark and the logo. If the respondents are successful in relation to either of its first two submissions, then the TCCC's trade mark infringement case must fail and TCCC accepts that that is the case. The respondents' third submission, assuming their first two submissions fail, is that they are using the whole shape of the Carolina Bottle as a trade mark and that the whole shape of the Carolina Bottle is not deceptively similar to any of TCCC's trade marks. Finally, the respondents submit that, even if they are using the outline or silhouette of the Carolina Bottle as a trade mark, that outline or silhouette is not deceptively similar to any of TCCC's trade marks.

Use as a trade mark 196 In Coca-Cola v All-Fect Distributors, the Full Court said (at 115): ... Use "as a trade mark" is use of the mark as a "badge of origin" in the sense that it indicates a connection in the course of trade between goods and the person who applies the mark to the goods: see Johnson & Johnson Australia Pty Ltd v Sterling Pharmaceuticals Pty Ltd (1991) 30 FCR 326 at 341, 351. 197 In The Shell Company of Australia Limited v Esso Standard Oil (Australia) Limited (1963) 109 CLR 407 ("Shell v Esso Standard Oil"), the principal question before the High Court was whether the defendant had used an animated "oil drop" cartoon character advertising its petrol in advertising films as a trade mark. The Court answered that question in the negative. Kitto J delivered the leading judgment for the majority. His Honour said that whether a sign is being used as a trade mark depends on how the viewer would see the particular use in the setting or context which applies. His Honour described the question as being (at 425): … whether, in the setting in which the particular pictures referred to were presented, they would have appeared to the television viewer as possessing the character of devices, or brands, which the appellant was using or proposing to use in relation to petrol for the purpose of indicating, or so as to indicate, a connexion in the course of trade between the petrol and the appellant. Did they appear to be thrown on to the screen as being marks for distinguishing Shell petrol from other petrol in the course of trade? 198 Kitto J also made the point that whether a sign or mark is being used as a trade mark requires a consideration of the purpose and nature of the use (at 426). 199 The need to consider the setting or context in determining whether there has been the use of a sign or mark as a trade mark is illustrated by the approach taken by Sundberg J in Global Brand Marketing Inc v YD Pty Ltd (2008) 76 IPR 161; [2008] FCA 605 ("Global Brand") at 166-168. In that case, his Honour was considering trade marks in relation to footwear and he referred to the marks in issue before him as the shape mark and sole mark. One of the questions was whether the respondent was using the shape or sole pattern of its shoes as a trade mark. In identifying the relevant context or setting for the determination of that issue, Sundberg J addressed and made findings about the relevant market and the shoe industry, including the nature of the outlets from which the shoes were sold, the price at which the applicant's shoes and the respondent's shoes were sold, how the applicant's shoes and the respondent's shoes were displayed for sale, the characteristics of the class of consumers likely to buy the applicant's shoes and the respondent's shoes, including whether the shoes were purchased after a visual inspection and a fitting. I am entitled to have regard to my own experience in making findings about these issues, provided that I am alert to the danger of allowing my own idiosyncratic knowledge or temperament to influence my decision (at 168, [30]-[31] per Sundberg J). 200 In Johnson & Johnson Australia Pty Limited v Sterling Pharmaceuticals Pty Limited (1991) 30 FCR 326 ("Johnson & Johnson") (at 347-349), Gummow J identified a number of principles relevant to whether there has been use as a trade mark in the context of a case involving a word mark. First, he said that a court is entitled to take into account a usage common in the relevant trade. Secondly, he said that the court is entitled to consider the way in which the word has been displayed in relation to the goods or advertisement of which complaint is made. Thirdly, his Honour noted that a word may have a descriptive element, but still serve as a badge of origin. That proposition can, I think, be adapted to the shape of a container. A bottle for non-alcoholic beverages contains otherwise formless goods and, therefore, one of the functions it performs is as a container. The fact that a bottle performs that function does not, of itself, mean that the shape of the bottle cannot operate as a trade mark. Furthermore, the fact that the shape of a bottle includes a decorative element does not mean the shape may not be used as a trade mark (see Adidas AG and Another v Pacific Brands Footwear Pty Ltd (No 3) (2013) 103 IPR 521; [2013] FCA 905 ("Adidas AG v Pacific Brands Footwear")). Fourthly, his Honour said that there might be trade mark use and infringement even though a defendant adds words to indicate that it, rather than the plaintiff, is the trade origin of the goods or services in question (see Saville Perfumery Ld v June Perfect Ld and F.W. Woolworth & Co Ld (1941) 58 RPC 147 ("Saville Perfumery v June Perfect") at 161 per Lord Greene MR). Finally, his Honour said that the fundamental question is whether "those to whom the user is directed" are being invited to purchase the goods of the defendants which are to be distinguished from the goods of other traders partly because they are described by the words in question. His Honour emphasised the words "partly because" and noted that this formulation of the fundamental question is based on the statements of Williams J in Mark Foy's Limited v Davies Coop and Company Limited and Another (1956) 95 CLR 190 at 205. 201 In Global Brand, Sundberg J identified some general principles in relation to shape marks which are, with respect, helpful. First, his Honour said that whether a person has used a shape or a feature of a shape as a trade mark is a matter for the court, and cannot be governed by the absence of evidence on the point. Secondly, his Honour said that context "is all important" and will typically characterise the mark's use, either trade mark use or not. Thirdly, his Honour said that there is a spectrum involving, at one end, shape or features that are purely functional and, at the other end, features of a mark that are non-descriptive and non-functional. They may make a shape "more arresting of appearance and more attractive, thus providing a means of distinguishing the goods from those of others" (at 175, [62]-[63]). 202 I have already described the context or setting in which the Coca-Cola beverages in the Contour Bottle and the Pepsi beverages in the Carolina Bottle are offered for sale and sold (at [182]-[187]). 203 As I have already said, the goods in this case are the non-alcoholic beverage which is formless. The particular issues which arise where it is alleged that the goods themselves are being used as a trade mark do not directly arise in this case. Those issues, and the cases in which they have been considered, were discussed by Sundberg J in Global Brand at 172-175, [50]-[61] (see Koninklijke Philips v Remington (1999) at 183, [47] per Lehane J; Koninklijke Philips v Remington (2000) at 103-104, [16]-[17] per Burchett J (with whom Hill and Branson JJ agreed)). In this case, it is the shape of the container which is said to constitute use as a trade mark. 204 As I have said, there are two ways in which TCCC puts its case. First, it submits that the respondents used the bottle shape, that is to say, the whole bottle shape of the Carolina Bottle as a trade mark. Secondly, and TCCC put this as "separately and in addition to the overall bottle shape", it submits that the respondents used the outline or silhouette of the Carolina Bottle as a trade mark. 205 I start with the first way in which TCCC put its case. TCCC refers to a number of matters in support of its submission and I refer to each of these matters and state my conclusions in relation to them. 206 First, TCCC submits that consumers are familiar with traders using special shapes of beverage containers as an indicator of trade origin. It was submitted that this was established by TCCC's extensive and long running promotion of its famous Contour Bottle, as well as the sale of the bottle over many years, and by the varying and distinct shapes of a number of the third party bottles. While I accept that there has been extensive advertising and sales of the Contour Bottle by TCCC and that consumers associate it with TCCC, there is no evidence before me of the reaction of consumers to the Carolina Bottle or, more generally, of the reaction of consumers to bottle shapes. I do not accept that the different shapes of a number of the third party bottles and the additional bottles means, of itself, that consumers consider each different shape to be an indicator of trade origin. TCCC referred to the decision in Adidas AG v Pacific Brands Footwear, but I think that case is distinguishable because there was evidence which the Court accepted that signs as identifiers were generally placed where the signs said to be used as a trade mark were found (at 534-535). The evidence in this case does not support a conclusion that a consumer would consider that, every time a bottle of a different shape was produced, the shape of the bottle was being used as a badge of origin. 207 Secondly, TCCC submits that the bottle shape of the Carolina Bottle was being used as a trade mark because it was different from the shape of a generic or industry bottle by reason of features that attract the eye and render the bottle distinctive. There was evidence before me that there is a generic or industry bottle for non-alcoholic beverages and that, in broad terms, it is cylindrical in the bottom half and a conical neck above the shoulder. One of the third party bottles was identified as an example. It is the "Bravo Lightly Carbonated Blood Orange Drink". The generic or industry bottle does not have a waist. TCCC emphasised that feature but I think that, whilst that is a point of difference, it does not of itself make the Carolina Bottle distinctive. It was not unknown or unique for a bottle to have a waist. 208 Thirdly, TCCC points to the evidence of Mr Le Bras-Brown that PepsiCo set out to produce a unique and distinctive "proprietary" bottle and believed that it had achieved that aim. It wanted to design a bottle that was "unique" and "ownable", and a bottle that could be described as "uniquely Pepsi". Mr Le Bras-Brown said that emotionally he wanted his bottle design to embody and express the meaning of the Pepsi brand, rather than to resemble a different brand. He said that the bottle had to look different from the competition. In Australia, the Carolina design was referred to as "a new, unique and Pepsi-Cola proprietary contour 'Carolina' design". In answer to these submissions, the respondents submit that this evidence does not assist TCCC because the question is how consumers view the shape of the Carolina Bottle, and not what PepsiCo sought to achieve in designing it. Furthermore, they submit that some of the phrases used by Mr Le Bras-Brown were suggestive of matters which are relevant under designs legislation, rather than whether or not shape was being used as a badge of origin. The respondents' submissions have force, but I do not think that they mean that what PepsiCo considered it had produced is irrelevant. It is a matter to be taken into account. After all, PepsiCo is a major player in the market with very sophisticated design, marketing, and sales departments. 209 Fourthly, TCCC submits that PepsiCo underwent a similar design process in relation to its Carolina PET bottle and it also believed that, in the case of that bottle, it had produced a unique and distinctive "proprietary" bottle. I make the same comments in relation to this submission as I did in the case of the previous submission. The consideration is, however, relatively neutral because it relates to a different bottle. 210 Fifthly, TCCC relies on the fact that PepsiCo has registered various bottle shapes as trade marks in the United States. TCCC submits that this was "strongly indicative of [PepsiCo] implementing a business strategy whereby its non-generic soft-drink bottles were used as trade marks". I think that the short answer to this submission is that advanced by the respondents. No attempt has been made by the respondents to register the shape of the Carolina Bottle as a trade mark anywhere in the world and, in those circumstances, I think the factor identified by TCCC is neutral. I would not go as far as to accept the respondents' submission that it may be inferred from the fact that there has been no attempt to register the Carolina Bottle as a trade mark that the respondents did not intend to use the shape of the Carolina Bottle as a trade mark. 211 Finally, TCCC submits that the fact that the name and logo branding is only on one side of the Carolina Bottle, and occupies a relatively small proportion of the surface area of the bottle, suggests a trade mark purpose for the bottle shape that is independent of the name and logo branding. I have examined the photographs of the third party bottles to see whether it is common for the name and logo to be on only one side of the bottle. That exercise did not help a great deal because the photographs show only one side of the bottle. The bottles which I have been able to examine from all sides, other than the Contour Bottle and the Carolina Bottle, suggest that often a word mark or logo or both appears on more than one part of the bottle. That is not always the case and I do not think the evidence is sufficiently strong to enable me to draw a firm conclusion about the matter. 212 TCCC also submits that the argument put against it by the respondents that, if bottle shape is relevant for trade mark purposes, it is as part of a trade mark which includes the names and logos on the bottle is illogical because it means bottle shape is part of a trade mark when seen with name and logos, but not when those matters are obscured. TCCC's submission has a good deal of force. 213 In my opinion, in the context in which it is sold, the Carolina Bottle is distinctive and consumers would associate it with the Pepsi brand. That was the aim of PepsiCo in designing it and I think it achieved what it set out to do. It is the overall shape of the Carolina Bottle which is distinctive and an important aspect of that shape is the horizontal waves on the bottle. I mention that latter aspect because, as will be seen when I come to consider deceptive similarity, there is force in the respondents' submission that TCCC seeks to rely on the horizontal waves feature for its use as a trade mark argument, but ignore it for the purposes of deceptive similarity. The overall shape of the Carolina Bottle is being used by the respondents as a trade mark. 214 TCCC also submits that the outline or silhouette of the Carolina Bottle is used as a trade mark separately and in addition to the overall bottle shape. Again, I will identify each of the matters relied on by TCCC and state my conclusions in relation to them. 215 First, TCCC relies on Mr Crone's opinion that the most significant element of the appearance of the Carolina Bottle from a design perspective is "the overall contour shape highlighted by the low waisted shape with its thinnest diameter located in the lower section of the bottle as seen in profile". I do not accept that opinion. The question for the Court is whether consumers are being invited to purchase the respondents' beverages which are to be distinguished by the consumer from the goods of other traders partly because of the silhouette or outline of the Carolina Bottle. Put another way, the question is whether the Court can be satisfied that the outline or silhouette would appear to the consumer to be a sign the respondents were using or proposing to use to indicate a connection in the course of trade between the beverage and the respondents. In my opinion, the answer to that question is no. The fact that the outline or silhouette of a bottle may be one of the first matters that can be seen by a consumer from a distance, or may be the only feature that can be seen from a distance under poor visibility conditions (for example, where there is condensation in a refrigerator or the bottles have their word marks or logos facing away from the consumer), does not mean that the outline or silhouette is used as a trade mark. All bottles have an outline or silhouette and the fact that a bottle has a waist is not so extraordinary as to lead to the conclusion that that feature alone is being used as a trade mark. 216 Secondly, TCCC relies on the fact that, when in a refrigerator or cooler, the two dimensional outline or silhouette of the Carolina Bottle is both "readily observable and eye-catching, relative to other aspects of the shape of the bottle". I make three observations about that submission. First, I accept the evidence of Professor Klein that photographs will not always reflect what the human eye can see and, as far as the planograms are concerned, whilst they might reflect the manufacturer's preferred location of the bottles in a refrigerator or cooler, they may not reflect how the refrigerator or cooler will appear in a retail outlet. Secondly, depending on the lighting, whilst the outline or silhouette of the respondents' Carolina Bottle may be seen, the same may be said about a number of other bottles shown in the photographs or planograms. Thirdly, the outline or silhouette of the Carolina Bottle is likely to become less important in the consumer's mind as he or she approaches the refrigerator or cooler and focuses on word marks, logos, and brands. As I have said, the fact that an aspect of a product may be seen at one point does not lead to the conclusion that consumers would see it as a badge of origin. 217 Thirdly, TCCC relies on the importance that PepsiCo placed on the outline or silhouette of the Carolina Bottle during the design process for the bottle. It submits that, during the design process for the Carolina Bottle, the bottle was often drawn only in outline. I do not think that is correct. The drawings to which I was referred include the feature of the horizontal waves. Whilst it is true that PepsiCo in its internal correspondence and marketing materials referred to the silhouette as an important aspect of the bottle's shape, it also emphasised the horizontal wave feature. It is appropriate to take these matters into account, however, they are far from decisive because what the respondents consider a notable feature of the Carolina Bottle is not necessarily what the consumer sees as a feature indicating origin. 218 Finally, TCCC relies on the fact that the trade advertising PepsiCo Australia and Schweppes undertook in relation to the Carolina Bottle emphasised its outline or silhouette. It is true that the trade flyer in evidence emphasises the outline or silhouette of the bottle, although the feature of the horizontal waves can also be seen. However, this is a trade flyer, and the question is what the consumer would see as a badge of origin. It is clear from Mr Van Dijk's evidence that none of the respondents have engaged in any large-scale public advertising campaigns involving the Carolina Bottle. 219 Having regard to all the evidence, I am not satisfied that consumers would consider the outline or silhouette of the Carolina Bottle alone as a feature being used to indicate a connection between the Pepsi beverages and PepsiCo and PepsiCo Australia. The outline or silhouette is, of course, part of the overall bottle shape of the Carolina Bottle.

Deceptive similarity 220 The principles relating to deceptive similarity are well established and they were not in dispute. 221 In assessing deceptive similarity, the Court does not place the plaintiff's mark and the defendant's mark side by side. The comparison is between the impression or recollection of the plaintiff's mark by a person of ordinary intelligence and memory, and the defendant's mark: Australian Woollen Mills Limited v F S Walton & Company Limited (1937) 58 CLR 641 ("Australian Woollen Mills v Walton") at 658 per Dixon J (as his Honour then was) and McTiernan J; Shell v Esso Standard Oil at 415 per Windeyer J. 222 It is well-established that an intention on the part of the alleged infringer to appropriate part of the trade or reputation of a rival may be sufficient in a borderline case to establish deceptive similarity: Australian Woollen Mills v Walton at 657 per Dixon and McTiernan JJ. I will discuss the relevant principle in more detail in the context of the passing off and Trade Practices Act causes of action. 223 The usual manner in which ordinary people behave is the test of what, if any, confusion or deception may be expected. On the one hand, potential buyers of the goods are not to be credited with any high perception or habitual caution, and, on the other hand, exceptional carelessness or stupidity is to be disregarded. The context, including the type of goods, the way they are displayed, and the circumstances in which they are sold, provides the standard. Often the question of deceptive similarity is not susceptible of much discussion and depends on a combination of visual impression and judicial estimation of the effect likely to be produced in the course of the ordinary conduct of affairs (Australian Woollen Mills v Walton at 658-659 per Dixon and McTiernan JJ). 224 Bearing in mind the matters referred to in the previous paragraph, the Court must make allowance for imperfect recollection of the plaintiff's mark (In the Matter of an Application by Rysta Ld to register a Trade Mark (1943) 60 RPC 87 at 108 per Luxmoore LJ). 225 The plaintiff's mark may be infringed by a mark which takes an essential feature of the mark. In De Cordova and Others v Vick Chemical Company (1951) 1B IPR 496; (1951) 68 RPC 103, Lord Radcliffe delivering the advice of the Privy Council said (at 498-499): Their Lordships consider that the Court of Appeal were right in holding that the appellants had infringed trade mark No 1852. They have not used the mark itself on the goods that they have sold, but a mark is infringed by another trader if, even without using the whole of it upon or in connection with his goods, he uses one or more of its essential features. The identification of an essential feature depends partly on the court's own judgment and partly on the burden of the evidence that is placed before it. … The likelihood of confusion or deception in such cases is not disproved by placing the two marks side by side and demonstrating how small is the chance of error in any customer who places his order for goods with both the marks clearly before him, for orders are not placed, or are often not placed, under such conditions. It is more useful to observe that in most persons the eye is not an accurate recorder of visual detail, and that marks are remembered rather by general impressions or by some significant detail than by any photographic recollection of the whole. 226 Lord Greene MR in Saville Perfumery v June Perfect at 162, referred to the fact that traders' customers often do not carry in their head the details of particular marks: In such cases the mark comes to be remembered by some feature in it which strikes the eye and fixes itself in the recollection. Such a feature is referred to sometimes as the distinguishing feature, sometimes as the essential feature, of the mark. 227 Although an essential feature of a mark is sufficient, in this case TCCC identified what it contended was the essential feature of the mark. In my discussion, I will refer to the essential feature of the mark. 228 The main or general idea of a mark seems to be a similar concept to that of the essential feature of the mark (see Jafferjee v Scarlett (1937) 57 CLR 115, at 121-122 per Latham CJ). 229 It is to be remembered that everything depends on the facts of the particular case, and a mark may not involve simply one essential feature (Crazy Ron's Communications Pty Ltd and Others v Mobileworld Communications Pty Ltd and Others (2004) 61 IPR 212; [2004] FCAFC 196 at 235, [102]-[103]). 230 A mark is deceptively similar to another mark if a number of consumers will be caused to wonder whether it might not be the case that the two products come from the same source. This test is different from that applied in a passing off action. In Southern Cross Refrigerating Company v Toowoomba Foundry Proprietary Limited (1954) 91 CLR 592, the High Court said (at 608): ... It is, of course, for the person applying for registration to establish that there is no likelihood of confusion and we agree with Kitto J. that registration should be refused if it appears that there is a real risk that "the result of the user of the mark will be that a number of persons will be caused to wonder whether it might not be the case that the two products came from the same source"; it is, of course, not necessary that it should appear that the user of the mark will lead to passing-off: see per Morton J. (as he then was) in In re Hack's Application. ... (Citation omitted). 231 I turn now to apply these principles to the facts of this case.

TM 63697 and TM 767355 232 It is convenient to deal with TM 63697 and TM 767355 together. 233 The features of TM 63697 were identified by the Full Court in Coca-Cola v All-Fect Distributors Ltd at 123 as follows: (1) the sides of the bottle are curved, rather than flat; (2) there is fluting on the top and lower portions of the bottle and no fluting in a central section; (3) the top and lower portions of the bottle have the same number of flutes; and (4) the bottle has a flat base and a banded neck. 234 Although I am not bound by that description, it accords with my own view of the mark and I adopt it. I think that a similar description can be given to TM 767355 with the addition that the fluting is more clearly shown and the scalloping can be seen in the representations for TM 767355. 235 The Carolina Bottle does not have flutes or a clear belt band, and it has the horizontal wave feature. In addition, its waist is more gradual and appears to extend higher up the bottle. On the face of it, these differences are significant. 236 TCCC seeks to overcome the weight which might otherwise be placed on those differences by submitting that the essential feature or the dominant feature or the main idea of TM 63697 and TM 767355 is the outline or silhouette of the bottle depicted in those marks. In one sense, TCCC narrowed that proposition even further by arguing that the essential feature was the overall low waisted contoured shape. In support of that argument, it relied on three matters. First, it pointed to Mr Crone's evidence that the trade marks reflect the overall low waisted contoured shape with its proportions as the single most recognisable element of the appearance of the TCCC Contour Bottle from a design perspective. Secondly, it relied on the fact that consumers will have an imperfect recollection of the marks and will not remember the other details. Thirdly, it argued that in assessing the question of deceptive similarity, the Court is entitled to consider all the surrounding circumstances, including the fact that consumers will see the bottles when they are filled with brown cola making other features, such as the flutes and the horizontal waves, much less noticeable (see In the Matter of an Application by the Pianotist Company Ld for the Registration of a Trade Mark (1906) 23 RPC 774 at 777 per Parker J). 237 I do not accept the submission that the outline or silhouette is the essential feature of TM 63697 or TM 767355. 238 I take into account the context or surrounding circumstances in which cola products are likely to be purchased as set out above (at [182]-[187]). As far as TM 63697 is concerned, the vertical flutes and the clear belt band are, to my mind, as prominent as the outline or silhouette. I see no reason to give any feature prominence over the other features. It cannot be said, for example, that a bottle with a waist is so extraordinary, or a bottle with vertical flutes and a clear belt band so common, that the outline or silhouette should be considered the essential feature. 239 These conclusions also apply to TM 767355. Having regard to the representations, the position is even stronger in relation to this trade mark. The perspective view is a view looking down at the bottle and shows the bottle as a three dimensional object. It does not show the outline or silhouette of the bottle at eye level. From the plan view, one can see the scalloping and its juxtaposition with the clear belt band. Both the typical elevation and the perspective view show the fluting far more clearly than it is shown in TM 63697. I do not think that it can be said that the essential feature of TM 767355 is the outline or silhouette. 240 TCCC puts its argument in another way. It submits that, even if I reject its argument that the respondents are using the outline or silhouette of the Carolina Bottle as a trade mark (which I do), nevertheless, in the context of the use of the shape of the Carolina Bottle as a trade mark, the overall impression of the Carolina Bottle is of a bottle having a low waisted contoured shape. That was the opinion of Mr Crone "from a design perspective". I do not accept that that is the view which would be held by the ordinary consumer. In my opinion, the waist, the horizontal wave feature, and, to a lesser extent, the frustoconical neck are the significant features of the Carolina Bottle. 241 As I have said, TCCC also submits, insofar as it is a borderline case, that the respondents had an intention to deceive or an intention to copy the Contour Bottle. For the reasons already given, I do not think the first respondent, in designing the Carolina Bottle, intended to appropriate part of TCCC's trade or reputation by deceiving or confusing consumers, and TCCC's reliance on such an argument must be rejected. In any event, even if such an intention had been established, it would only be relevant in a borderline case and I do not consider this to be such a case. 242 The appropriate comparison is between all of the features shown in TM 63697 and TM 767355 (including, for example, the vertical fluting), and not just their outline or silhouette, on the one hand, and the whole of the Carolina Bottle on the other. The shape of the Carolina Bottle is not deceptively similar to TM 63697 or TM 767355. Even if I am wrong and the appropriate comparison is between the outlines or silhouettes of TM 63697 and TM 767355 on the one hand, and the outline or silhouette of the Carolina Bottle on the other, I do not think that there is deceptive similarity for the reasons given in the context of TM 1160893 and TM 1160894 (see [248] below).

TM 1160893 and TM 1160894 243 It is also convenient to deal with TM 1160893 and TM 1160894 together as they raise similar issues. 244 In the case of both of these marks (as well as TM 767355), there is a statement that the provisions of s 41(5) of the Trade Marks Act applied. At the time of registration of those marks, that subsection dealt with cases where, on an application for registration, the Registrar of Trade Marks concludes that the trade mark is to some extent inherently adapted to distinguish the relevant goods or services from the goods or services of other persons, but is unable to decide, on that basis alone, that the trade mark meets the requirement for registration of being capable of distinguishing the applicant's goods or services from the goods or services of other persons. The Registrar is then required to perform an exercise which may lead to the conclusion that the trade mark does or will distinguish the relevant goods or services as being those of the applicant for registration. 245 The representations of these marks are not without their difficulties. In the case of both marks, there seems to be more of a bulge in the middle section on the right hand side than on the left hand side. Furthermore, TM 1160894 is fuzzy around the outline and it is difficult to know what is depicted in the cap area. In fact, it is not entirely clear from the representation alone that it is a bottle, although that is probably clear when regard is had to the class of goods with respect to which it is registered. 246 Nevertheless, there are relatively few difficulties in construing these trade marks when they are applied or used as a drawing or picture. Their use, at least in a fairly general way, may be seen in TCCC's advertising material which was identified by Ms Davis. As a picture or drawing, the marks may be part of advertising brochures or pamphlets, or may be images on the television, or they may be applied to an advertising medium such as an advertising billboard or even the side of a bus. They may be placed on the goods themselves and examples of that (although without the black outline) may be seen in the placing of what appears to be TM 1160893, again in a fairly general way, on cans containing the Coca-Cola beverage. I say in a fairly general way because one can see differences in the contoured shape on close examination (see [44] above). 247 However, the allegation in this case is that the marks have been infringed by the production of a three dimensional object (i.e., a bottle) that has a deceptively similar outline or silhouette to the registered marks. TCCC submits that the registered marks are really the outlines or silhouettes, and that a fair notional use of the enclosed area is for labels and other marks. In the alternative, it submits that that feature (i.e., the outlines or silhouettes) is the essential feature or main idea of the marks. As I understand it, the respondents submit that the white and black infill are features of the marks. It is not easy to construe the registered marks when one is considering three dimensional objects. As a consumer would know, bottles in the marketplace do not look like TM 1160893 or TM 1160894. I am disposed to think that TCCC's submission is correct, but I do not need to decide the point. Even if the outline or silhouette is the only feature of the marks, or is the essential idea of the marks, the comparison is with the sign the alleged infringer has used as a trade mark. In this case, I have found that is the whole shape of the Carolina Bottle. The following are the distinctive features of the Carolina Bottle which I think are distinctive but are not part of the registered marks: (1) the Carolina Bottle has a gently curving waist at a higher point than that in the marks and does not have an abrupt pinch near the base; (2) the Carolina Bottle has a cylindrical shoulder, not a curved shoulder; (3) the Carolina Bottle has a frustoconical neck, not a curved neck; (4) the Carolina Bottle has a twist top enclosure, not a cap lid seal; and (5) the Carolina Bottle has a distinctive horizontal embossed wave pattern across the bottom half of the bottle. 248 The first four of the five features involve the outline or silhouette of the Carolina Bottle and, it seems to me that, even if I was wrong and the relevant comparison is between the silhouettes or outlines of TM 1160893 and TM 1160894 on the one hand and the silhouette or outline of the Carolina Bottle on the other, nevertheless, the outline or silhouette of the Carolina Bottle would not be deceptively similar to either TM 1160893 or TM 1160894. 249 I do not think the Carolina Bottle is deceptively similar to TM 1160893 or TM 1160894.

Conclusion on Trade Mark Infringement 250 TCCC's case on trade mark infringement fails.

passing Off and the CLAIMS UNDER THE Trade Practices ACT

Relevant Principles 251 The three core concepts in the law of passing off are reputation (in the relevant mark or other indicia), misrepresentation (leading to deception), and damage to goodwill (or the likelihood thereof): Conagra Inc v McCain Foods (Aust) Pty Ltd (1992) 33 FCR 302 ("Conagra Inc v McCain Foods") at 356 per Gummow J. 252 In Reckitt and Colman Products Ltd v Borden Inc and Others (1990) 17 IPR 1, Lord Oliver of Aylmerton described the three elements in the following terms (at 7): ... First, he must establish a goodwill or reputation attached to the goods or services which he supplies in the mind of the purchasing public by association with the identifying get-up (whether it consists simply of a brand name or a trade description, or the individual features of labelling or packaging) under which his particular goods or services are offered to the public, such that the get-up is recognised by the public as distinctive specifically of the plaintiff's goods or services. Secondly, he must demonstrate a misrepresentation by the defendant to the public (whether or not intentional) leading or likely to lead the public to believe that goods or services offered by him are the goods or services of the plaintiff. Whether the public is aware of the plaintiff's identity as the manufacturer or supplier of the goods or services is immaterial, as long as they are identified with a particular source which is in fact the plaintiff. For example, if the public is accustomed to rely upon a particular brand name in purchasing goods of a particular description, it matters not at all that there is little or no public awareness of the identity of the proprietor of the brand name. Thirdly, he must demonstrate that he suffers or, in a quia timet action, that he is likely to suffer damage by reason of the erroneous belief engendered by the defendant's misrepresentation that the source of the defendant's goods or services is the same as the source of those offered by the plaintiff. (see also TGI Friday's Australia Pty Ltd and Another v TGI Fridays Inc and Another (1999) 45 IPR 43; [1999] FCA 304). 253 In the ordinary case, the applicant's reputation is determined at the time the respondent introduces his goods into the market. However, in this case TCCC argues for a different approach. It argues that the quantity of sales of the Carolina Bottle between August 2007 and May 2008 was very small and that, in May 2008, it was withdrawn from the market and not reintroduced into the market until February 2009. TCCC argued that the relevant date for assessing its reputation in the Contour Bottle was February 2009, not August 2007. 254 The reason the applicant's reputation is assessed at the date that the respondent introduces his goods into the market is that, if there is no passing off at that stage, there cannot be at a later stage because the applicant cannot increase its reputation in the mark or other indicia in the face of the respondent's mark or indicia. To take an example of a distinctive shape, if the applicant did not have sufficient reputation in the shape when the respondent introduced his shape into the market, then he cannot establish a reputation thereafter and at the same time argue that consumers will be deceived into thinking the goods come from the same source, or there is an association, by the respondent's shape. Similar reasoning applies to the alleged contraventions of the Trade Practices Act. 255 In effect, TCCC argues that the quantity of sales of the Carolina Bottle before February 2009 was so small as to be de minimus and that, in the circumstances, it should be ignored. 256 In Thai World Import & Export Co Ltd and Another v Shuey Shing Pty Ltd and Others (1989) 17 IPR 289, Thai World claimed passing off and misleading or deceptive conduct by Shuey Shing in 1988 and thereafter. It claimed a reputation in supplying a peeled mung bean product in a blue packet which was damaged by Shuey Shing supplying its peeled mung bean product in a blue packet. However, Thai World had withdrawn its blue packet in 1984 and supplied its peeled mung bean product in a green packet from 1984 to 1988. It reintroduced its blue packet in 1988. It failed in its claim because the cause of any deception in the market place in 1988 was the re-entry onto the market of its blue packet, rather than any conduct on the part of Shuey Shing. Gummow J said (at 303): In the light of the evidence, I also accept that any goodwill and reputation in Victoria and South Australia in relation to the marking and get-up of the Thai World blue packet had faded some time before 1988. That desuetude is the result of the switch by Thai World to the green packet coupled with Thai World's delay with knowledge, in pursuing any rights it may have had upon becoming aware (some time between 1984 and 1985) of sales of the Sun World blue packet. Thai World took a decision at that time to do nothing because, as Mr Kimson put it, the quantity and the money involved were not then very great, but that was a significant omission on Thai World's part. To adapt what was said by Deane and Fitzgerald JJ in Taco Co of Australia Inc v Taco Bell Pty Ltd (1982) 42 ALR 177 at 204, the consequence of this loss of reputation is that when in 1988 the Thai World blue packet was reintroduced into Victoria and South Australia in some numbers, any likelihood of deception could not properly be seen as the result of conduct of Shuey Shing in continuing its by then well established trade in the Sun World blue packet. The cause of any such likely deception would not be any conduct for which Shuey Shing was responsible; rather it would be the re-entry upon the market of the Thai World blue packet: cf Parkdale Custombuilt Furniture Pty Ltd v Puxu Pty Ltd (1982) 149 CLR 191 at 199, per Gibbs CJ; 42 ALR 1. 257 In Optical 88 Ltd v Optical 88 Pty Ltd and Others (No 2) (2010) 89 IPR 457; [2010] FCA 1380, Yates J considered a different aspect of the issue. In that case, the applicant sought to bolster its reputation by relying on its conduct from the date upon which the contravening conduct first started and the date of commencement of proceedings. Yates J rejected the argument saying (at 520): … I cannot see how any after-acquired reputation can assist it in the face of what would otherwise have been the legitimate commencement and continued conduct of the first respondent's business. 258 Where an applicant alleges that a respondent has engaged in passing off by using a trade mark or similar indicium, he must show that his own trade mark or indicium is known in the relevant market and the reputation associated with the mark is, to some degree, substantial. Passing off includes a representation by the respondent that leads a consumer or customer to believe that the respondent's business is associated with the applicant's business (Erven Warnick BV v J Townend & Sons (Hull) Ltd [1979] AC 731, at 741-742 per Lord Diplock). 259 The tort of passing off is complete upon the making of the representation and relief may be granted even if any deception or erroneous belief in the mind of a consumer or customer is dispelled before sale: Caltex Oil Australia P/L v S & M Motor Repairs Pty Ltd (1987) AIPC 90-931. That means that in a case such as the present the tort will be complete even if on reaching the cash register the consumer realises his mistake and does not proceed with the purchase. The same applies to the misleading or deceptive conduct claim. The fact that a purchase does not proceed will be relevant to damages, not liability. At the same time, it is important to note that confusion by itself is not enough to establish passing off or misleading or deceptive conduct. 260 The plaintiff in a passing off action need not prove an actual subjective intention to mislead: Sydneywide Distributors v Red Bull Australia Pty Ltd at 372 per Weinberg and Dowsett JJ (with whom Branson J agreed). However, it is well-established that proof of an actual subjective intention to mislead will be evidence in a borderline case that the defendant's conduct has misled or deceived its audience: Australian Woollen Mills v Walton at 657 per Dixon and McTiernan JJ. A distinction must be drawn between an intention to copy and an intention to deceive. For the principle identified by Dixon and McTiernan JJ in Australian Woollen Mills v Walton to be engaged, an intention to deceive must be shown, and an intention to copy does not always indicate an intention to deceive: Conagra Inc v McCain Foods at 345 per Lockhart J; Apand Pty Limited v The Kettle Chip Company Pty Limited (1994) 52 FCR 474 at 495. 261 As far as the Trade Practices Act claims are concerned, the court considers the likely effect of the respondents' conduct on "ordinary" or "reasonable" members of the class of prospective purchasers of (in this case) cola products, and is entitled to put to one side reactions which may be considered extreme or fanciful (Campomar Sociedad, Limitada and Another v Nike International Limited and Another (2000) 202 CLR 45 ("Campomar v Nike") at 86). 262 In the case of misleading or deceptive conduct, or conduct likely to mislead or deceive, it is not sufficient for the applicant to establish that a consumer has been caused to wonder whether, for example, the two products have come from the same source. Actual deception or the likelihood of actual deception must be proved. "Likelihood" in this context includes a real possibility. Furthermore, all of the circumstances, including any word marks, device marks, or logos on the accused bottle, must be taken into account. These propositions are well-established (Parkdale Custom Built Furniture Proprietary Limited v Puxu Proprietary Limited (1982) 149 CLR 191 at 199 per Gibbs CJ; at 209-210 per Mason J (as his Honour then was); at 225-226 per Brennan J (as his Honour then was); Campomar v Nike at 84, [100]; Butcher and Another v Lachlan Elder Realty Pty Limited (2004) 218 CLR 592 at 605, [39] per Gleeson CJ, Hayne and Heydon JJ; at 625-626 [109] and [112] per McHugh J). 263 Therefore, the causes of action for passing off and for contraventions of the Trade Practices Act differ from the cause of action for trade mark infringement in two respects. First, they require proof of actual deception, or the likelihood of actual deception, and a finding of cause to wonder is not sufficient. Secondly, in addition to a consideration of the sign found to be used as a trade mark (i.e., in this case, the whole shape of the Carolina Bottle), these causes of action require a consideration of the effect of the word marks and device marks on the Contour Bottle and the Carolina Bottle. The authorities establish that, depending on circumstances, the causes of action may be made out even though the products have different brand names on them. That will occur where the get-up of the products is so similar that the Court finds that deception will, or is likely to, result. Examples of where the Court has so found are WD & HO Wills (Australia) Ltd v Philip Morris Ltd (1997) 39 IPR 356 (a s 52 case); Sydneywide Distributors v Red Bull Australia (a passing off and a s 52 case); Peter Bodum A/S and Others v DKSH Australia Pty Ltd (2011) 92 IPR 222; [2011] FCAFC 98 at 273-275 [255]-[265] per Greenwood J (with whom Tracey J agreed) (a passing off and a s 52 case).

Application of Principles 264 There is no dispute that TCCC has a reputation in its brand name and the words written in Spencerian script. The same, I think, can be said of PepsiCo and its brand name and the globe device. 265 The first question is whether TCCC has a reputation in the Contour Bottle and, if so, whether it is a reputation in the overall shape of the bottle or is a reputation in the outline or silhouette by itself. 266 As to the date upon which TCCC's reputation is to be assessed, there is no precise evidence about the number of Carolina Bottles which were offered for sale or sold between August 2007 and May 2008 other than that it was a small, perhaps a very small, number. As I have said, the Pepsi beverages in the Carolina Bottle have not been the subject of any advertising to consumers. In the circumstances, the matter is largely one of impression. I doubt that the introduction of the Pepsi beverages into the Australian market had any substantial impact in the market before the reintroduction of the Carolina Bottle in February 2009. That suggests that the proper date for the assessment of TCCC's reputation is February 2009. 267 The difference between the two dates does not, however, have any practical relevance because, although I think TCCC had a reputation in the overall shape of the Contour Bottle by August 2007, I do not think TCCC had a reputation in the outline or silhouette of the Contour Bottle alone in either August 2007 or February 2009. Another way of putting this point is that, although TCCC's advertising and promotion of the Contour Bottle between those dates continued to include the outline or silhouette of the Contour Bottle, it did not result in a change in TCCC's reputation in the Contour Bottle. 268 In August 2007, TCCC had a reputation in the Contour Bottle as a whole. I mean by that every aspect of the shape and features of the bottle which would be noticed by the ordinary consumer. In my view, that includes the low pinched waist and the fluting. Other features which may be less important but which, I think, would be noticed by the ordinary consumer are the clear belt band and the particular curvature of the neck and shoulder. I think it likely that only a very observant consumer would notice the scalloping. 269 It seems to me that it is one thing to say that a consumer would regard a bottle with these features as a TCCC bottle. It is another to say that he or she would make a decision to purchase having regard to these features alone, that is to say, without regard to the word mark and other branding. Professor Klein thought that a consumer would probably not do that. She considered that a consumer was most likely to rely on brand names and logos as the most valid cue. In other words, as Professor Klein said, the shape of a bottle might be memorable without necessarily being a diagnostic cue of origin. Professor Klein did acknowledge that her "opinion" had not been verified by empirical studies. Nevertheless, I think that there is a good deal of force in Professor Klein's view that the consumer would treat the branding on the bottle as the most valid cue. After all, this is not a case where the manufacturer of the product is not well-known, or the labelling is not prominently displayed or is in some way difficult to access. For example, this case does not involve a label on a container next to goods displayed out of their container, or a label or tag underneath a sofa or chair. Furthermore, the brands "Coca-Cola" and "Pepsi" are very well-known, as are their word marks and device marks. The word marks and device marks are prominently displayed on the bottles for the purpose of identifying the source of the product. Furthermore, I accept, as was said in the course of evidence, that the words "Coca-Cola" in Spencerian script would be one of the world's most recognisable advertising images. In all of those circumstances, it is difficult to see why an ordinary consumer would not make his or her purchasing decision by reference to the brand names, device marks, or logos. 270 Despite the force of these matters, and not without some hesitation, I think that a sufficient number of consumers may well make their selection based on overall bottle shape and I have in mind the consumer who self selects from the refrigerator with minimal involvement in the purchase (see [185] above). 271 The difficulty for TCCC is that, even accepting that and accepting that both bottles will contain dark brown cola and be sold within a similar, if not the same, context, I do not think that such a consumer would be misled or deceived, or would be likely to be misled or deceived, in the case of overall bottle shape because I think he or she would detect quite clearly the difference between the Contour Bottle and the Carolina Bottle. The most noticeable difference between the two bottles is that the Contour Bottle has the very distinctive fluting and the Carolina Bottle has the distinctive horizontal waves. Other noticeable features are the different shaped neck and shoulders and the fact that the waist on the Contour Bottle is lower and more pinched. In other words, if overall bottle shape is the cue, I do not think that there is any real likelihood of deception. 272 It will be apparent from what I have said that I am not satisfied that TCCC had a reputation in the mere outline or silhouette of the Contour Bottle in August 2007 or February 2009. TCCC did use the outline or silhouette in its advertising and I think the ordinary consumer would probably recognise the bottle once his or her attention was directed to the word "Coke" appearing inside the silhouette, or the words "Coca-Cola" appearing nearby. However, I do not think that the nature or degree of advertising was such as to suggest that, at either point in time, consumers were so familiar with the mere outline or silhouette as to automatically associate it with TCCC's beverages. Nor, as I have said, was there evidence that consumers in Australia treated, or were accustomed to treat, the outline or silhouette of bottles containing non-alcoholic beverages alone as indicators of trade origin. 273 The context in which Coca-Cola and Pepsi beverages are offered for sale and sold, which I have outlined above, does not persuade me that a consumer selects his or her cola drink by reference to outline or silhouette alone. Once the other features of shape are taken into account, there is no possibility of deception. 274 Although these conclusions stand by themselves, I note that they are supported by the practical circumstances that there is no evidence of any empirical tests suggesting that consumers have been misled by the Carolina Bottle or, although not a strong point, of any complaints to PepsiCo Australia or Schweppes about the Carolina Bottle and any similarity it bears to the Contour Bottle. The evidence of Ms Le of the one occasion where she momentarily mistook a Carolina Bottle for the Contour Bottle is insufficient to lead to a different conclusion in light of all the matters I have identified. I am disposed to think that the mistake came about at a point before Ms Le had processed all the information that would be processed by the ordinary consumer. 275 In order to deal with all the arguments, I should mention that the respondents relied on two practical matters in support of the proposition that TCCC, as applicant and a major supplier in the market for the supply of cola products, did not consider that the respondents had been guilty of passing off or of contraventions of the Trade Practices Act. First, the respondents pointed to the fact that the Carolina Bottle was reintroduced into the Australian market in February 2009 and TCCC became aware of the presence of the Carolina Bottle in the market in early 2010. TCCC did not institute this proceeding until October 2010. Some of this delay may be explained by the low sales of the Carolina Bottle and the lack of consumer advertising with respect to the bottle. In any event, I do not consider the delay to be so significant as to support the inference for which the respondents contend. Secondly, the respondents pointed to the fact that, although the Carolina Bottle has been introduced into a number of countries around the world, TCCC has only taken proceedings of a similar nature to this proceeding in Australia, Germany, and New Zealand. There might be many reasons for this related to the particular markets and legal systems of other countries and TCCC's operations in those countries. To explain these matters might involve a very broad ranging inquiry and I do not think TCCC should be criticised for not undertaking that inquiry. 276 The respondents made reference to the fact that TCCC had not adduced any evidence of complaints or inquiries made to it about the alleged similarities between its Contour Bottle and the Carolina Bottle. This is correct and I think it is a matter to be noted in the same way as I have noted that PepsiCo Australia and Schweppes have not received any complaints or inquiries. It may be explained by the low sales of the Carolina Bottle and the possibility that some consumers realise their error before purchase and do not consider it worthwhile to complain. As I have been able to reach a clear conclusion without regard to this matter, and as I do not consider it to be of great significance, even if relevant, I leave it out of account.

Conclusion on Passing Off and Contraventions of the Trade Practices Act 277 For the reasons I have given, the claims in passing off and for contraventions of the Trade Practices Act must be dismissed.

conclusion 278 TCCC's application must be dismissed. I certify that the preceding two hundred and seventy-eight (278) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Besanko.