CATCHWORDS CUSTOMS AND EXCISE - classification - whether finding of fact by the Administrative Appeals Tribunal that goods were 'paper' rather than 'nonwoven' supportable on the evidence - whether subheading referring to "aseptic paper" extended to goods substantially composed of such material. Customs Tariff Act 1987 (Cth), Sched 3. Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 Chinese Food and Wine Supplies Pty Ltd v Collector of Customs (Vic) (1987) 72 ALR 591 Collector of Customs v Pozzolanic Enterprises Pty Ltd (1993) 43 FCR 280 Rheem Australia Ltd v Collector of Customs (NSW) (1988) 4 ALD 615 Sharp Corporation of Australia Pty Ltd v Collector of Customs (1995) 59 FCR 6 Vernon-Carus Australia Pty Ltd v Collector of Customs (1995) 21 AAR 450 COLLECTOR OF CUSTOMS v JOHNSON & JOHNSON MEDICAL PTY LTD NG 725 of 1995 Sackville J. Sydney 4 September, 1996 IN THE FEDERAL COURT OF AUSTRALIA) NEW SOUTH WALES DISTRICT REGISTRY) No. NG 725 of 1995 GENERAL DIVISION ) On appeal from the General Administrative Division of the Administrative Appeals Tribunal constituted by Mr M.D. Allen (Senior Member), Professor G.A.R. Johnston (Member), and Mr I.R. Way (Member). BETWEEN: COLLECTOR OF CUSTOMS Applicant AND: JOHNSON & JOHNSON MEDICAL PTY LTD Respondent CORAM: SACKVILLE J. PLACE: SYDNEY DATE: 4 SEPTEMBER, 1996 MINUTES OF ORDER THE COURT ORDERS THAT: 1. The appeal be dismissed, with costs. NOTE: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules. IN THE FEDERAL COURT OF AUSTRALIA) NEW SOUTH WALES DISTRICT REGISTRY) No. NG 725 of 1995 GENERAL DIVISION ) On appeal from the General Administrative Division of the Administrative Appeals Tribunal constituted by Mr M.D. Allen (Senior Member), Professor G.A.R. Johnston (Member), and Mr I.R. Way (Member). BETWEEN: COLLECTOR OF CUSTOMS Applicant AND: JOHNSON & JOHNSON MEDICAL PTY LTD Respondent CORAM: SACKVILLE J. PLACE: SYDNEY DATE: 4 SEPTEMBER 1996 REASONS FOR JUDGMENT Introduction This is an appeal by the Collector of Customs (the "Collector") pursuant to s.44 of the Administrative Appeals Tribunal Act 1975 (Cth) from a decision of the Administrative Appeals Tribunal (the "AAT"), given on 23 August 1995. The AAT's decision related to the classification of goods, which it described as surgical drapes and surgical packs, for the purpose of Schedule 3 to the Customs Tariff Act 1987 (Cth) (the "Tariff Act"). The goods were imported into Australia by the respondent ("Johnson & Johnson"). The AAT decided that the goods should be classified within item 4823.90.10 of Schedule 3 of the Tariff Act (as it stood at the relevant time), namely: "Good as follows: (a) aseptic paper". Item 4823 is within Chapter 48 of Schedule 3. The title to Chapter 48 is "Paper and paperboard; articles of paper pulp, of paper or of paperboard". The Collector's principal contention is that the goods should be classified within Chapter 63, the title to which is "Other made up textile articles; sets; worn clothing and worn textile articles; rags". The Collector says the appropriate classification is item 6307.90.49 ("Other"), which is within a sub-heading "Nonwovens" (item 6307.90.4). If this contention is incorrect, the Collector submits as an alternative argument that the goods should be classified to item 4818.90.00, the terms of which are set out below. If the AAT's classification is upheld, the goods are free of duty; if either of the Collector's contentions is correct, duty is payable at the rate of 15%. An extract from Textile Terms and Definitions (9th ed), published by the Textile Institute of the United Kingdom, was in evidence before the AAT. The extract is useful because it defines (at 211-212) the word "nonwoven" and assists in understanding the issues before the AAT. The definition commences as follows: "nonwoven; nonwoven fabric n. or adj. Opinions vary as to the range of fabrics to be classified as nonwovens. In general, they can be defined as textile structures made directly from fibre rather than yarn. These fabrics are normally made from continuous filaments or from fibre webs or batts strengthened by bonding using various techniques: these include adhesive bonding, mechanical interlocking by needling or fluid jet entanglement, thermal bonding and stitch bonding... The controversial areas are: (1) wet-laid fabrics, containing wood pulp, in which the boundary with paper is not clear, (ii) stitch-bonded fabrics, which contain some yarn for bonding purposes; (iii) needled fabrics containing reinforcing fabric". The Tariff Act The structure of the Tariff Act and of Schedule 3 to the Act has recently been helpfully described by Northrop J. in Vernon-Carus Australia Pty Ltd v Collector of Customs (1995) 21 AAR 450 (FCA/FC). I have borrowed from his Honour's outline in setting out the relevant provisions. The Tariff Act imposes duties of customs on goods imported into Australia: s.21(1)(a). The rate of duty in respect of goods (where the goods are not derived from a "Preference Country") is ascertained by reference to the general rate set out in the third column of the tariff classification under which the goods are classified: s.22(a). The Act provides that the "Interpretation Rules" must be used for ascertaining the tariff classification under which goods are classified: s.10. The "Interpretation Rules" are defined to mean the General Rules for the Interpretation of the Harmonized System provided for by the International Convention on the Harmonized Commodity Description and Coding System, done at Brussels on 14 June 1983: s.5. The English text of the General Rules is set out in Schedule 2 to the Act. In Vernon-Carus v Collector, Northrop J. described (at 451) Schedule 3 as "very lengthy and [containing] very complex provisions". The Schedule is divided into a number of Sections, Chapters and sub-Chapters. Within each Chapter there are a number of headings and sub-headings. In general, the heading provides a generic classification of goods. The rate of duty is specified with respect to particular types or classes of goods identified in sub-headings. Rule 1 of the General Rules provides as follows: "The titles of Sections, Chapters and sub-Chapters are provided for ease of reference only, for legal purposes, classifications shall be determined according to the terms of the headings and any relative Section or Chapter Notes and, provided such headings or Notes do not otherwise require, according to the following provisions". The "following provisions" are those specified in rules 2-6, the relevant terms of which are extracted later. It follows from rule 1 that the titles of Sections and Chapters are provided for ease of reference only. The Headings Chapter 48 is within Section X of Schedule 3. The title to Section X is as follows: "Pulp of wood or of other fibrous cellulosic material; waste and scrap of paper or paperboard; paper and paperboard and articles thereof". I have already referred to the title of Chapter 48. The following headings and sub-headings of Chapter 48 are relevant: "Chapter 48 4818 TOILET PAPER, HANDKERCHIEFS, CLEANSING TISSUES, TOWELS, TABLECLOTHS, SERVIETTES, NAPKINS FOR BABIES, TAMPONS, BED SHEETS AND SIMILAR HOUSEHOLD, SANITARY OR HOSPITAL ARTICLES, ARTICLES OF APPAREL AND CLOTHING ACCESSORIES, OF PAPER PULP, PAPER, CELLULOSE WADDING OR WEBS OF CELLULOSE FIBRES: ... 4818.90.00 - Other. ... 4823 OTHER PAPER, PAPERBOARD, CELLULOSE WADDING AND WEBS OF CELLULOSE FIBRES, CUT TO SIZE OR SHAPE; OTHER ARTICLES OF PAPER PULP, PAPER, PAPERBOARD, CELLULOSE WADDING OR WEBS OF CELLULOSE FIBRES ... 4823.90.10 --- Goods, as follows: (a) aseptic paper; ... 4823.90.90 ---- Other." I have already referred to the title of Chapter 63. Chapter 63 is within Section XI, the title to which is "Textiles and Textile Articles". The following headings and sub-headings of Chapter 63 are relevant: "6307 OTHER MADE UP ARTICLES, INCLUDING DRESS PATTERNS: 6307.90.4 --- Nonwovens: 6307.90.49 --- Other." The Notes to Section XI provide as follows: "1. - This Section does not cover: ... (m) Products or articles of Chapter 48 (for example, cellulose wadding)". The General Rules I have already cited the terms of rule 1 of the General Rules. The remaining Rules relevant to this case are as follows: "2. (a) Any reference in a heading to an article shall be taken to include a reference to that article incomplete or unfinished, provided that, as presented, the incomplete or unfinished article has the essential character of the complete or finished article. It shall also be taken to include a reference to that article complete or finished (or falling to be classified as complete or finished by virtue of this Rule), presented unassembled or disassembled. (b) Any reference in a heading to a material or substance shall be taken to include a reference to mixtures or combinations of that material or substance with other materials or substances. Any reference to goods of a given material or substance shall be taken to include a reference to goods consisting wholly or partly of such material or substance. The classification of goods consisting of more than one material or substance shall be according to the principles of Rule 3. 3. When by application of Rule 2(b) or for any other reason, goods are, prima facie, classifiable under two or more headings, classification shall be effected as follows: (a) The heading which provides the most specific description shall be preferred to headings providing a more general description. However, when two or more headings each refer to part only of the materials or substances contained in mixed or composite goods or to part only of the items in a set put up for retail sale, those headings are to be regarded as equally specific in relation to those goods, even if one of them gives a more complete or precise description of the goods. (b) Mixtures, composite goods consisting of different materials or made up of different components, and goods put up in sets for retail sale, which cannot be classified by reference to 3(a), shall be classified as if they consisted of the material or component which gives them their essential character, insofar as this criterion is applicable. (c) When goods cannot be classified by reference to 3(a) or 3(b), they shall be classified under the heading which occurs last in numerical order among those which equally merit consideration. ... 6. For legal purposes, the classification of goods in the subheadings of a heading shall be determined according to the terms of those subheadings and any related Subheading Notes and, mutatis mutandis, to the above Rules, on the understanding that only subheadings at the same level are comparable. For the purposes of the Rule the relative Section and Chapter Notes also apply, unless the context otherwise requires." History of the Case This is the second occasion in which the classification of the goods in suit has been before the Federal Court. Indeed, the review process commenced as long ago as 16 August 1991, when Johnson & Johnson sought review of decisions by the Collector that the goods imported into Australia should be classified to item 4818.90 of Schedule 3 to the Tariff Act. On 8 June 1993 the AAT (comprising Senior Member Mr M.D. Allen and Members, Professor G.A.R. Johnston and Mr I.R. Way) set aside the Collector's decision and remitted the matter, with a direction that the goods were to be classified to item 4823.90.10. The Collector appealed against the AAT's decision. The appeal was determined by Einfeld J., who gave judgment on 27 July 1994. His Honour set aside the AAT's decision and remitted the matter for a rehearing. His Honour observed that the AAT had not attempted to define "paper", although he also noted that the only expert evidence had been given by Ms Moller, a testing controller from the Wool Testing Authority, who did not claim expertise in relation to paper. Nonetheless, Ms Moller had concluded that the fabric of which the goods were principally comprised was a non-woven fabric. While the AAT rejected Ms Moller's evidence, Einfeld J. considered that it had misunderstood Ms Moller's evidence and had therefore rejected the evidence on an erroneous basis. It followed that the AAT's conclusion that the fabric was "paper" (rather than nonwoven) was "contrary to the evidence". The matter was remitted to the AAT and a further hearing took place before the AAT, comprising the same three members. At that further hearing, the parties relied on the evidence in the first AAT proceedings, but also adduced further evidence. The AAT gave its decision on 23 August 1995. The AAT reached the same conclusion it had in the first proceedings and made the same orders. The present appeal is brought from the AAT's decision of 23 August 1995. The Goods In its first decision, the AAT noted that the goods had been described in Appendices 1 and 2 to the application. The AAT continued as follows: "3....Those goods in Appendix 1 were identified as surgical drapes, whereas those in Appendix 2 were surgical packs, ie a series of surgical gowns and specifically cut drapes for particular surgical procedures put up in sets. The surgical drapes listed in Appendix 1 and the gowns and some of the drapes in the surgical packs as listed in Appendix 2, were made from a fabric with the trade name of Fabric 450, although they were in places reinforced with materials made of other substances. 4. The surgical packs also contained specifically cut drapes made from non-woven fabric other than Fabric 450 and other minor items such as paper hand towels. The purpose of the packs was to deliver into an operating theatre a sterile pack of disposable protective drapes and clothing. The purpose of the drapes (Appendix 1) and the items in the packs (Appendix 2) was to provide sterile covering materials for patients, theatre staff and equipment, in order to prevent a cross flow of infection between equipment, theatre staff and the patient, particularly with regard to body fluids. Following the operation the drapes and clothing can then be disposed of as opposed to the more traditional linen drapes which are laundered. Such disposable drapes etc have been increasingly used since the increased numbers of patients suffering from AIDS and Hepatitis B. 5. Mr Herbranson is an employee of the Applicant with particular knowledge of the subject goods. He stated that Fabric 450 is made of a blend of paper pulp and polyester using a hydroentanglement process to "entangle" the paper pulp into the polyester web. The mixture is 55% paper pulp to 45% polyester and the percentages of materials are tightly controlled in the manufacturing process. After the fabric is produced by the hydroentanglement process it is put through a repelling treatment to give it the ability to repel fluids. After the subject goods are put up into packs, those packs are then sterilised by a cobalt irradiation process." This description was repeated in the second AAT decision and its accuracy was not disputed by the parties. The evidence before the AAT provided more detailed information on the components of the goods. For example, one of the surgical packs was a cystoscopy pack, which consisted of the following: "Component Materials 1 X Abdominal T Sheet Fabric 450 Drisite Microcide Cotton Gauze 2 X Leggings Fabric 450 2 X O.R. Towel Wood pulp Nylon Scrim 1 X Overwrap/Tablecover Anti-static polyethylene Wood pulp" The Second Decision In its second decision, after repeating the description of the goods, the AAT identified the witnesses who had given evidence at the second hearing and referred to their respective experience and fields of expertise. These witnesses were: l Mr P. Fuller, Johnson & Johnson's Director of Marketing; l Mr L.S. Budden, Johnson & Johnson's Development Manager; l Mr R. T. Fish, a consultant to the paper industry (called as a witness by Johnson & Johnson); and l Mr R.W. Maddern, also a consultant to the paper industry (called by the Collector). The AAT referred to note 1(m) to Section XI of Schedule 3 to the Tariff Act, pointing out that this is a note that "otherwise require[s]" within rule 1 of the General Rules: Liebert Corporation Australia Pty Ltd v Collector of Customs, (FCA/FC, 1 November 1993, unreported), at 18-20. It followed that, before considering the question of whether the goods could be classified under a heading or sub-heading within Chapter 63, it was necessary positively to exclude their classification under some heading of Chapter 48. The AAT then addressed the composition of Fabric 450. "11.No evidence was adduced in these proceedings to challenge the earlier evidence that Fabric 450 is made by the "hydroentanglement" process and is in the proportion of 55% paper pulp to 45% polyester web by weight. 12.As was pointed out by Mr Herbranson in the original proceedings, the paper pulp in Fabric 450 comes from a very large roll of paper tissue which is unrolled and laid down over a web of polyester that has been put onto a moving rack and then that goes through what was termed the hydroentanglement process. 13.The rolls of paper tissue are manufactured by traditional paper making technologies, namely, a non-woven wet laid process. There is no dispute that these roles [sic] are paper consisting of cellulose fibres. These cellulose fibres make up 55% by weight of the final Fabric 450. 14.The chemical constitution of cellulose fibres is such that hydrogen bonding may occur between cellulose fibres so as to bind them together. The traditional method of wet laying of cellulose fibres to produce paper products results in hydrogen bonding between the cellulose fibres.... 15.Cellulose fibres and polyester fibres have inherently different chemical natures so that they do not naturally bond together. Hydrogen bonds do not normally form between cellulose fibres and polyester fibres. 16.As was pointed out by Mr Budden in his evidence, the cellulose fibres and polyester fibres in Fabric 450 are mechanically bonded together by the hydroentanglement process. 17.The hydroentanglement process utilises the energy of water to entangle the cellulose fibres into the web of polyester fibres. The process may be described by saying that there is a sheet of polyester attached to the tissue paper, like a laminate, and then the hydroentanglement process creates a physical linking of fibres together (or intermingling) so that it stays together. This process does not mean that the process is destroyed as a paper." After referring to some further evidence, the AAT continued: "22.The micrographs provide no evidence that there is any chemical interaction between the cellulose fibres and the polyester fibres in Fabric 450 to make them a different product. The microscopic evidence is that the cellulose fibres in Fabric 450 behave as do the cellulose fibres in Kleenex tissues. Both the cellulose fibres and the polyester fibres are clearly identifiable as such under the microscope. 23.The two types of fibres in Fabric 450 are physically interlaced by the hydroentanglement process. The essential properties on each type of fibre are apparent in the final product. The process is that traditionally prepared tissue paper is physically bonded with 45% by weight of polyester fibres in a hydroentanglement process. The process results in a final product in which two layers of fibres are discernible under the microscope. One layer is predominantly cellulose fibres laid out as in tissue paper but containing some polyester fibres forced into this layer by the hydroentanglement process - these polyester fibres crossing both layers thus providing the mechanical bonding between the two types of fibres. 24.Apart from the scientific evidence detailed above as to manufacture and composition of Fabric 450 the Tribunal notes that the said product exhibits many of the properties of paper, namely, absorption, porosity, softness, feel and the stopping of bacteria" The AAT found that Fabric 450 was a mixture, within the meaning of rule 2(b). By reason of note 1(m) to Section XI, rule 3 had no application to the case. Accordingly, references to "paper" in headings and sub-headings within Chapter 48 included references to a product comprised of paper plus other materials. According to the AAT, both Mr Fish and Mr Maddern (the latter reluctantly) had accepted the International Standards Organisation ("ISO") definition of paper, which states that paper can be manufactured from a mixture of cellulosic and synthetic fibres. The AAT considered that Ms Moller's evidence was unhelpful, since the definitions on which she relied did not define the boundary between a nonwoven fabric and paper. Moreover, her expertise was in textiles and her evidence was of little value in distinguishing between paper and nonwovens. In addition, her tests were based on a mistaken assumption that Fabric 450 was produced by a wet-laid process, whereas in fact it had been produced by the hydroentanglement process. The AAT expressed its finding this way: "31.Having regard to the scientific evidence, the properties of the goods and the applicable Interpretative Rules, the Tribunal finds that Fabric 450 is a paper, and is thus to be classified to Chapter 48." The AAT then noted that there had been nothing in the material adduced to cause it to alter its earlier finding that the purpose of the goods is to create a sterile field and prevent cross-infection. Having regard to the goods themselves and what they were designed to do, it was clear that they were not similar to any other goods "within the genus of Heading 4818". None of these goods maintained a sterile field during surgery and none tended to repel fluids. Sub-heading 4823.90.10 expressly provided for the condition of asepsis, the specific purpose for which the goods were designed. Finally, the AAT rejected the Collector's argument that the drapes and surgical packs could not be classified under item 4823.90.10, since they were articles made up from aseptic paper. The Collector had argued that the aseptic paper could not be classified as goods, but was merely a component from which the goods (drapes and surgical packs) were made. The AAT rejected this submission because, in its view, the word "goods" should be given its ordinary English meaning of "articles of trade" or "merchandise". The word therefore covered the paper of which the drapes and packs were largely made. The Collector's Submissions Mr Roberts, on behalf of the Collector, submitted that it was erroneous for the AAT to take any course other than to identify Fabric 450 as a nonwoven. As I followed his argument, Mr Roberts accepted that the AAT had not departed from the approach dictated by cases such as Re Gissing and the Collector of Customs (1977) 14 ALR 555 (AAT), at 556-557 and Collector of Customs v Chemark Services Pty Ltd (1993) 114 ALR 531 (FCA/FC), at 534. Those cases require the classifier to identify the goods objectively and then match the identification with a heading or sub-heading in Schedule 3. In this case the AAT had identified the goods as surgical drapes and surgical packs. No heading or sub-heading specifically referred to surgical drapes or packs. It was therefore open to the AAT to approach the task of identification by taking into account the materials or substances from which the goods were manufactured: Re Tridon Pty Ltd and Collector of Customs (1982) 4 ALD 615 (AAT), at 620-621. Thus the AAT could ascertain the principal material from which the drapes and most of the items comprising the surgical packs were made, that is, Fabric 450. The flaw identified by Mr Roberts in the AAT's reasoning was that all the evidence pointed to Fabric 450 being a nonwoven fabric and not paper. Mr Roberts went so far as to contend that it had never been an issue that Fabric 450 was anything other than a nonwoven. He submitted that, having regard to the evidence, the AAT erred in making any finding other than that Fabric 450 was a nonwoven. It followed that the AAT was bound to classify the goods within a heading or sub-heading that specifically refers to nonwovens and, of the relevant headings, the most appropriate was item 6307.90.49. Mr Roberts put his submissions in two ways, although in substance they seem to amount to the same point. First, he said that the AAT had erred by making a critical finding of fact - that Fabric 450 was a paper rather than a nonwoven - without any evidence to support the finding and, indeed, in the face of uncontradicted evidence that Fabric 450 was a nonwoven. As he pointed out, it is an error of law to make a finding of fact which has no probative evidence to support it or to draw inferences which are not reasonably open on the facts: Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321, at 359-360, per Mason CJ.; Federal Commissioner of Taxation v McCabe (1990) 26 FCR 437 (FCA/Davies J.), at 438-442. The second way in which Mr Roberts put the point was that the AAT erred in applying note 1(m) to Section XI to exclude the operation of Chapter 63 of Schedule 3. The error, as I followed the argument, was said to arise because note 1(m) could be relevant only if the subject goods were paper or of paper. Since the uncontradicted evidence was that Fabric 450, the principal component of the goods, was a nonwoven fabric, the goods were not capable of falling within Chapter 48 and, accordingly, note 1(m) could not apply. This way of putting the argument, as Mr Roberts acknowledged, depended upon the proposition that it was not reasonably open to the AAT to make the finding that Fabric 450 was a paper. I did not understand Mr Roberts to dispute that, if it were reasonably open to the AAT to make the finding, note 1(m) was applicable to the case. In support of his submissions, Mr Roberts took me through much of the evidence before the AAT. He pointed out that a number of witnesses had given evidence that they would regard Fabric 450 as a nonwoven. Mr Herbranson, for example, was a Marketing Director of Johnson & Johnson. He acknowledged in his evidence that he would describe Fabric 450, produced by the spun-lace (or hydroentanglement) method, as a nonwoven fabric. Mr Burnes was a manufacturer, whose products included hospital uniforms, hospital theatre products and general ward linen items. Mr Burnes was familiar with Fabric 450 and said that he would describe it as a "nonwoven product". An article in the trade magazine "NonWovens Industry" referred to surgical packs and gowns manufactured, inter alia, by Johnson & Johnson as "nonwovens". Ms Moller, although mistaken in her assumption that Fabric 450 was produced by the wet-laid method (as opposed to hydroentanglement), had also given evidence that Fabric 450 was a nonwoven fabric. Mr Roberts also took me to the evidence of the expert witnesses who had not given evidence at the first hearing, but who did give evidence at the second hearing. He submitted that these witnesses provided no support for the finding made by the AAT, since they either regarded Fabric 450 as a nonwoven or were not prepared to describe it as paper. Mr Roberts referred to and relied on a statement made by Johnson & Johnson's legal representative at the first AAT hearing. This statement, which was made in the course of Mr Burnes' evidence, was as follows: "...there seems to be a lot of uncertainty as to where this case is heading. There is no dispute whatever. Our witness has said that these fabrics were non-woven fabrics. There is no dispute at all. What that means, legally, is quite a distinct question, but if the tribunal would feel assisted by the statement on behalf of the applicant that the fabric in question is a non-woven fabric, then I make that statement." Mr Roberts relied on this statement as demonstrating that it had never been in issue that Fabric 450 was a nonwoven. The Evidence In the light of the Collector's arguments, it is necessary to canvass the evidence before the AAT to ascertain whether it was reasonably open to it to find that Fabric 450 was paper. It is important to appreciate that the material before the AAT was not confined to the material adduced at the first hearing. It was on the basis of that material that Einfeld J. set aside the first decision. The principal difficulty facing the AAT was that no witness could identify a general test or standard that could be applied to distinguish paper from nonwoven articles and, in particular, to determine whether Fabric 450 is properly to be regarded as paper or a nonwoven. Mr Fish, whose expertise was in the paper industry, gave the following evidence: "And are you aware of anything - any standard, any criteria, or definition, or test in the paper industry that tells you when something stops being a paper and commences being a non-woven?---I am unaware of any - I have not encountered any criteria in use in the paper industry that attempts [sic] differentiation between paper and a non-woven fabric." Mr Budden, whose experience related both to textiles and fibrous products, gave this evidence: "But are you aware of any general test, or standard that one can apply to determine whether fabric 450 is paper or a nonwoven?---I'm not aware of any specific test that you can do which, if it meets a certain criteria one can say this is definitely a paper article, or this is definitely a nonwoven article. Now, would it be fair to say that different experts in the textile and nonwoven fields could test fabric 450 and come up with different conclusions as to whether it was a paper or nonwoven?---Yes, as I've just mentioned, there are different tests method requirements. For example tear tests, I understand that the tear test that is used within the paper industry is a different test, and it is a standard test in Australian standards, or ISO standards. The technique of the test is different to that used in the textile industry. So depending on which test method one used for the testing, for example, of tear characteristics of fabric 450, I believe one would get a different result." Mr Fish, upon whose evidence the AAT specifically relied, expressed the view that the most authoritative definition or description of "paper" was that published by the ISO: "paper: A generic term for a range of materials in the form of a coherent sheet or web - excluding sheets or laps of pulp as commonly understood for papermaking or dissolving purposes and non-woven products - made by deposition of vegetable, mineral, animal or synthetic fibres, or their mixtures, from a fluid suspension onto a suitable forming device, with or without the addition of other substances. They may be coated, impregnated or otherwise converted, during or after their manufacture, without necessarily losing their identity as paper. In conventional papermaking processes, the fluid medium is water; new developments, however, include the use of air and other fluids." As Mr Fish's evidence suggests, it follows from the ISO definition that paper: (a) can be made from vegetable or cellulosic fibre or synthetic fibres, or a mixture of both; (b) can be made with or without the addition of other substances; (c) can be coated, impregnated or otherwise converted, during or after manufacture without necessarily losing its identity as paper; and (d) is conventionally made by processes using water, but can be made by other processes, using air and fluids other than water. The difficulty of drawing the dividing line between paper (which, as all expert witnesses agreed, is made by a nonwoven process) and nonwovens is illustrated by the circularity of the definitions referred to in evidence. The ISO definition of "paper" excludes "non-wovens", but does not (yet) define the term "nonwovens". The standard definitions of "nonwoven" cited by the Textile Institute simply exclude "paper". The two definitions are as follows: "(1) Nonwovens are defined under ISO 9092:1988 as: 'A manufactured sheet, web or batt of directionally or randomly orientated fibres, bonded by friction, and/or cohesion and/or adhesion, excluding paper and products which are woven, knitted, tufted, stich-bonded incorporating binding yarns or filaments, or felted by wet-milling, whether or not additionally needled. The fibres may be of natural or man-made origin. They may be staple or continuous filaments or be formed in situ'. (2) Nonwoven fabric is defined under ASTM D 1117-80 as: 'A textile structure produced by bonding or interlocking of fibers, or both, accomplished by mechanical, chemical, thermal, or solvent means and combinations thereof. Discussion: The term does not include paper or fabrics that are woven, knitted, or tufted'." (Emphasis added.) Mr Fish gave evidence of the characteristics of paper. These included its surface strength, absorbency, porosity to air and steam, optical properties and feel. He also said that paper can have chemicals added to it to give it the quality of repellency; those chemicals might give the paper greater strength. The AAT specifically found that Fabric 450 exhibited many of the properties of paper, "namely, absorption, porosity, softness, feel and the stopping of bacteria". With the exception of the last characteristic (not mentioned by Mr Fish), the finding of the AAT is amply supported by the evidence. Mr Svehla, who appeared for Johnson & Johnson, filed supplementary written submissions which detailed the evidence supporting the findings and it is unnecessary to repeat the references here. (The AAT's reference to the "stopping of bacteria" appears to be intended to reflect Mr Budden's evidence that paper - which is normally permeable - can be constructed so as to provide a bacterial barrier.) The AAT also specifically found that Fabric 450 comprised 55%, by weight, of traditionally prepared tissue paper and 45%, by weight, of polyester fibres. This finding was supported by the evidence of Mr Herbranson and Mr Budden. Mr Budden said that, although the dividing line between paper and nonwovens was not clear cut, one test for distinguishing the two was the respective weight of the component fibres. The AAT was also entitled to take account of other evidence in reaching its findings. This included the following: l Mr Budden's evidence that the hydroentanglement process used to make Fabric 450 did not destroy the essential nature of the cellulose fibre; l Mr Budden's evidence that the hyroentanglement process used to make Fabric 450 disturbed less than 50% of the hydrogen bonding (a chemical process that causes attraction and adhesion between wood pulp fibres) in the cellulosic fibres; l Mr Maddern's concession that the cellulose was not chemically bonded to the synthetic fibres, but simply mechanically interlocked; and l Mr Maddern's concession that, although it took considerably more force to tear Fabric 450 than tissue paper, the tearing process produced similar consequences, in that the cellulose fibres were teased apart. Each of these items of evidence supported the AAT's findings. The AAT again expressly rejected the evidence of Ms Moller. It did so partly, but not wholly, on the basis that she had mistakenly assumed that Fabric 450 was made from a wet-laid process. It also took into account Ms Moller's lack of expertise concerning paper. In my opinion, the AAT was clearly entitled to take this course. In any event, it committed no error in rejecting her evidence. The AAT also implicitly rejected Mr Maddern's view that Fabric 450 was not paper. Mr Maddern gave evidence that, despite the ISO definition, he had never seen paper made from a mixture of cellulosic and synthetic fibres. He took the traditional view that paper comprises cellulose fibres bonded together with water. Mr Maddern accepted that he had not carried out tests to ascertain Fabric 450's qualities of absorption or repellency and that he had not specifically tested it for tearing strength. Clearly, it was necessary for the AAT to form a view about Mr Maddern's evidence. The members of the AAT had the opportunity to observe him giving evidence and to assess his approach. The question is not whether the AAT was correct in preferring Mr Fish's approach to that of Mr Maddern, but whether it was reasonably open to the AAT to take this course. In my view it was. Therefore no legal error has been shown in the course taken by the AAT. The AAT's reasons concentrated on the evidence of the expert witnesses given at the second hearing. There is nothing to suggest that the AAT ignored the views expressed by the non-expert witnesses, such as Mr Burnes and Mr Herbranson. The AAT was quite entitled to give their evidence relatively little weight, having regard to their expertise or lack of it. The AAT had the advantage of evidence from witnesses with greater expertise in the composition and manufacture of paper or nonwovens, or both. Mr Burnes, for example, acknowledged that he did not know how nonwoven fabrics were manufactured. Mr Herbranson, although able to describe in general terms the manufacturing process for Fabric 450, was principally involved in sales and marketing. Mr Roberts referred to Chinese Food and Wine Supplies Pty Ltd v Collector of Customs (Vic) (1987) 72 ALR 591 (FCA/FC). There, Lockhart J. (with whom Woodward and Ryan JJ. agreed) observed (at 599) that, in deciding whether particular imported goods fell within an item in Schedule 3, the characteristics of the goods, including get-up, colour, labelling and packaging are all relevant considerations. So much may be accepted. But the evidence from some witnesses that the goods in suit were not, for example, marketed or labelled as paper does not materially assist in deciding whether the factual conclusion reached by the Tribunal was reasonably open to it. Finally, I do not think the statement made by Johnson & Johnson's legal representative at the first hearing prevented the AAT reaching the conclusion it did. It is not clear what the legal representative had in mind in making the statement and Mr Svehla was unable to shed light on the question. However, the simple fact is that, after Einfeld J. set aside the first decision of the AAT, the matter was fought on the very issue of whether Fabric 450 was a paper or a nonwoven. Fresh evidence was called and neither the Collector nor the AAT considered that Johnson & Johnson was foreclosed from pursing the issue. In these circumstances, the statement of the legal representative does not itself, or in combination with other circumstances, render the AAT's findings unreasonable. In my opinion, the AAT could reasonably find in the material before it that Fabric 450 was paper. The case is not one where the AAT's decision can be described as perverse or lacking any evidentiary foundation. As I have already said, the question for the Court is not whether the AAT was correct in reaching the conclusion it did, nor whether other fact-finders might have taken a different approach. The AAT was faced with the difficult task of drawing a line between paper and nonwovens, when standard definitions or criteria offered relatively little guidance. It approached the task by attempting to describe the characteristics of paper (as Einfeld J's judgment suggested it should) and ascertaining whether Fabric 450 had these characteristics, or sufficient of them, to warrant being identified as paper. The AAT concluded that Fabric 450 did have sufficient of those characteristics to warrant being so identified. That finding was open on the evidence. The Alternative Submission The Collector's alternative argument was that, assuming Fabric 450 to be paper, the AAT had erred in classifying the goods under item 4823.90.10 rather than item 4818.90.00. The error was said to be that the AAT construed sub-heading 4823.90.10, "Goods, as follows: (a) aseptic paper", as referring to aseptic paper as a component of the surgical drapes and surgical packs. Mr Roberts submitted that item 4873.90.10 did not apply to goods made from aseptic paper, where those goods had a separate identity as surgical drapes or surgical packs. In my opinion, as a matter of construction, sub-heading 4823.90.10 is not confined to aseptic paper in a "pure" state, as distinct from an article constituted substantially, if not wholly, of aseptic paper. Heading 4823 refers both to paper and to articles "of...paper". It seems to me that the cryptic language of sub-heading 4823.90.10 is capable of encompassing both categories. There is no obvious reason why a narrow construction of the sub-heading should be adopted and Mr Roberts did not suggest one. In my view, it was open to the AAT to make a finding based on an assessment of whether Fabric 450 constituted a sufficiently substantial component, for example, of the surgical drapes to justify classifying the drapes as aseptic paper (assuming, as the AAT found, that the necessary quality of asepsis was present). Whether the AAT was or was not correct in making this finding is a question of fact: Collector of Customs v Pozzolanic Enterprises Pty Ltd (1993) 43 FCR 280 (FCA/FC), at 287-288. In any event, Mr Roberts' submission does not, in my view, pay sufficient regard to the way in which the case was conducted and to the findings of the AAT. The AAT specifically found that the surgical drapes gave the surgical packs, identified in Appendix 2, their "essential character" within rule 3(b) of the General Rules. This is a finding of fact (Sharp Corporation of Australia Pty Ltd v Collector of Customs (1995) 59 FCR 6 (FCA/FC), at 13-14) not challenged by the Collector. I have also found that the AAT did not commit any error of law in classifying Fabric 450, the material from which the surgical drapes were made, as paper. Accordingly, once the surgical drapes were classified to Chapter 48 as paper, it was open to the AAT, pursuant to rule 3(b), to classify the goods within Chapter 48 as if they consisted of the material that gave them their essential character. I did not understand Mr Roberts to dispute that, if the AAT could classify the goods by reference to that material, it was open to the AAT to classify them as aseptic paper. As the AAT noted, the goods were sterile and the fabric offered additional protection in that it contained a bactericidal chemical. The AAT also found that the surgical drapes themselves were made from Fabric 450, reinforced in places with materials made of other substances. In its first set of reasons, to which the AAT referred in its second reasons, the AAT found that rule 2(b) applied to the surgical drapes. I did not understand that finding to be disputed. The operation of the equivalent to rule 2(b) was described by Wilcox J. in Rheem Australia Ltd v Collector of Customs (NSW) (1988) 78 ALR 285 (FCA/FC), at 294: "The sub-rule applies to goods constituting a mixture or combination, one of the ingredients of which complies with a description in a particular item, or sub-item or paragraph. The sub-rule then extends the reference in the Schedule so as to include the mixture or combination. It is not necessary to the application of r 2(2) that the mixture or combination itself fall within the description, or any description, in Sch.3. This would be an unlikely even and, were it to happen, r 2(2) would be unnecessary. There would already be available a relevant category in Sch.3." Rule 2(b) thus extends sub-heading 4823.90.10 to apply to the combination constituting the surgical drapes, even though the sub-heading describes only one of the ingredients of the surgical drapes. Conclusion In my opinion, the appeal should be dismissed with costs. I certify that this and the preceding 28 pages are a true copy of the Reasons for Judgment of the Honourable Justice Sackville. Associate: Dated: 4 September, 1996 Heard: 15 August, 1996 Place: Sydney Decision: 4 September, 1996 Appearances: Mr P. Roberts, instructed by Australian Government Solicitor, appeared for the applicant. Mr J.T. Svehla, instructed by Michell Sillar, Solicitors, appeared for the respondent.