The appellant undertook not to publish in any form whatsoever any Gallup Poll results in respect of which the respondents have the copyright. That undertaking would I consider be broken by the publication of such results as a news item or in the course of an editorial, or in the remarks of some commentator be he a Minister of the Crown or anyone else. But what is it that must not be published? What are Gallup Poll results in respect of which the respondents have the copyright? That question was approached in the argument in two very different ways. One, that which counsel for the appellant urged, was to ask what is the result of a Gallup Poll: and to say that, if it be something that cannot be the subject of copyright, then the undertaking is on a false premiss, and having been given without admissions it creates no enforceable obligation. The other approach, that which counsel for the respondents took, was to say that the undertaking assumes that the result of a Gallup Poll is something in which there can be copyright, and that this controls the meaning of the expression "result of a Gallup Poll", restricting it to a document in which there is copyright and which can properly be described as such a "result". Either approach leads to difficulties because of the nature of literary copyright, the form of copyright involved in the question here. There can be no copyright in mere information, in news, however much time, skill and care was given to the gathering of it. There can only be copyright in the literary form in which the information is presented. It appears from the statement of claim that the respondents base their claim to copyright in what they there call "the results of public opinion polls" partly upon the terms of an agreement in writing made between themselves on 15th October 1958 and a supplementary agreement of 11th July 1961. It is alleged that, pursuant to these agreements, the respondent Morgan conducts public opinion polls and supplies "the results" thereof and a "written report thereon incorporating the results thereof" to the other respondent, Australian Public Opinion Polls (The Gallup Method). That company then provides copies of the report to certain newspapers which are its customers. The appellant is not one of those customers. The evidence establishes, I think, that the report as furnished to the newspapers could be the subject of copyright. But that is not the question in the present proceedings, for the respondents seek to distinguish between the report on the survey and the results of the survey. They seem to me to be in some uncertainty as to the nature and the source of the rights they assert. The respondent Morgan says, in his affidavit of 8th November 1961, that the second-named respondent, the company, had before the date of his agreement with it of 15th October 1958 "acquired and still then retained the sole and exclusive rights within the Commonwealth of Australia to the use of the names "The Gallup Method" and "Gallup Poll" and also to the use of the "Gallup Method of conducting public opinion polls" ". This may mean only that the public opinion surveys conducted in Australia by the second respondent are, according to the evidence given by Morgan in these proceedings, known as Gallup Polls. But the respondents apparently consider that they have "acquired" some proprietary right in the term "Gallup Poll" - which they emphasize by the regular use of capital letters "The Gallup Poll" - and have acquired an exclusive right to use in public opinion surveys a technique that they call "The Gallup Method". On what basis this could be so was not stated. I cannot help thinking that there is some misconception, and that it has coloured the respondents' view of their rights, and of the meaning of the word "results" in its context, and of what would constitute a breach of the appellant's undertaking. On the other hand, I cannot help thinking that much of the argument that we heard for the appellant placed too narrow a meaning on the expression "results of a poll". It was argued that this meant only the crude figures obtained by the enquiry, not even stated as percentages; that it could not mean any conclusion derived from those figures by any method of analysis or by comparison with other figures or other information. On the other hand, it was argued in effect that what is the result of a public opinion survey depends upon the purpose for which the survey was conducted. If the purpose were to ascertain whether there had been any change of public opinion on some topic since an earlier date, the result sought could only be arrived at after a comparison. However that may be, I do not think that the meaning of the expression "Gallup Poll results", as used in the undertaking, is to be found by considering those words in the abstract and regardless of the practices and terminology of the art of conducting public opinion surveys. These surveys are made by the skilful use of a sampling process. They are made for many purposes, including the prediction of election results. The evidence in this matter showed, and it is a matter of fairly general knowledge, that the "Gallup Poll" is the name by which the public opinion surveys made by the American Institute of Public Opinion are commonly known in the United States. The American Institute of Public Opinion was founded by Doctor George Gallup in, I think, 1935. The use of the term "Gallup Poll" to distinguish its surveys from those made by other poll-takers seems to be well established in the United States. But elsewhere the term "Gallup Poll" seems to be a generic description. At all events, no basis appeared on which that term can in Australia become in law the exclusive property of a particular local poll-taker or be restricted to any opinion surveys except perhaps those actually conducted in the United States by the American Institute of Public Opinion. There is now a considerable body of literature, both books and periodicals, dealing with public opinion surveys or polls. I have read some of this material, including some writings of Doctor Gallup. I have read too parts of Mr. Cantrill's book Public Opinion 1935-1946, which states that its aim is to present as many opinion poll results as possible in a convenient and useful form. I base no conclusion upon my reading, for the matter was not given in evidence before us. But when the suit is heard, and more complete evidence is given, some reference may be made to the use which writers on this subject make of such expressions as "the results" and "the findings" of public opinion polls. I express no opinion as to the correctness of the construction of the appellant's undertaking that Else-Mitchell J. adopted. I do not mean to say that it was erroneous. But it seems to me impossible to say that the undertaking was clear or that a breach of it was certainly established. I agree in the view that my brother Owen expresses in the judgment that he is about to deliver that on this ground, if on no other, the appeal should be allowed. I would allow the appeal and discharge the order of the Supreme Court and in lieu thereof order that the application for a writ of sequestration be dismissed with costs. This order would, of course, not prejudice any application that the respondents (the plaintiffs in suit number 1761 of 1961 in the Supreme Court of New South Wales in Equity) may make to that Court for an interlocutory injunction or any application to that Court to have the suit brought on for hearing at an early date.