Cullen v Welsbach Light Company of Australasia Ltd [1907] HCA 3; (1907) 4 CLR 990 (18 March 1907)

Cullen (Executor of Lascelles) Defendant, Appellant; and Welsbach Light Company of Australasia Ltd. Plaintiffs, Respondents.

This was an action for an infringement of a patent, No. 11,247 of Victoria, for a hood for incandescent gas burners. The claim in the specification is for "an incandescent gas burner consisting of pure oxide of thorium with a small proportion of one of the rare metals herein mentioned prepared as described." The proportion mentioned in the body of the specification was "a very small proportion, generally not exceeding one or two per cent. of the oxides of certain other of the rare metals, namely, uranium, cerium, or prazeodymium." The mode of preparation was described as dissolving nitrate of thorium in the purest possible condition, and adding 1 to 2 per cent. of solution of nitrate of one of the other metals, preferably uranium. Various defences were set up, of which it is only necessary to mention two—that the invention was not new, and that it was publicly used by others than the inventor in Victoria before the grant of the letters patent. Hodges J., before whom the action was tried, found as a fact upon the evidence that the prior user by others was proved, and decided that the patent was for that reason invalid. The Full Court arrived at a different conclusion upon the evidence, and being of opinion that the other objections were all invalid, allowed the appeal and gave judgment for the plaintiffs. The matter has been very fully and ably argued, and many points of great and general interest have been discussed, with which, however, in the view which I take of the facts, it is not necessary to deal at length.

Mr. Mitchell, for the appellant, submitted the broad proposition that, if the product of an invention is publicly sold by the inventor, that fact is conclusive proof that the invention has been given to the public, and that there is no exception to this rule, unless, possibly, in the case where the product is of such a nature that the mode of production cannot be discovered from it. There are, no doubt, reported cases in which the language of the learned Judges is wide enough to cover this proposition to its fullest extent. But with one possible exception: Wood v. Zimmer[1], which is not fully reported, so wide a proposition was not necessary for the decision of any of the cases. The invalidity of a patent which follows from what is sometimes called, indiscriminately, public user may depend upon either of two reasons: (1) If the public user, whatever that term means, has been by the inventor himself, the invention is no longer new. In that case the objection is want of novelty. (2). If the public user, whatever that term means, has been by others, the grant is forbidden by the Statute of Monopolies, which is incorporated in the definition of the term in the Victorian Patents Act 1890. The ground of the objection founded upon user by the inventor alone is, therefore, that the invention has been published to the world. And so the case is treated by the Courts in Hancock v. Somervell[2] (quoted in all the text books); Morgan v. Seaward[3]; Harwood v. Great Northern Railway Co.[4]; Humpherson v. Syer[5], and other cases. But it is manifest that, whether an invention has been published or disclosed to the world or not is a question of fact. I will afterwards consider the question whether the use by the public of an article from which no information can be obtained as to the nature of the process by which it is produced is a user by "others than the inventor" within the meaning of the Statute of Monopolies, and the further question whether the disclosure of means of knowledge is conclusive or sufficient evidence of disclosure of the invention. If the sale or use of the article produced gives in fact no information to anyone as to the nature of the invention it is hard to discover any principle on which it can be said that there has been a disclosure.

In my opinion a distinction must be taken between different classes of cases. It is settled that a patent may be granted for an invention of a new thing or contrivance, i.e., for a new product, or for a new process for producing a thing previously known, if the other necessary conditions exist. If (Case I.) the invention is for producing a thing already known, e.g., the metal aluminium, or common salt, by a new process, the sale of the product will give no information as to the nature of the invention. If (Case II.) the invention is of a new contrivance, such as the lock which was in question in Carpenter v. Smith[6], the sale or public use of it by the inventor would of itself disclose the nature of the invention. But there are intermediate cases. The product may (Case III.) be of such a nature that any person conversant with the subject, and applying the common knowledge of the time, could bring about the same result. Of this class of case the "Magnolia Metal" in question in In re Miller's Patent[7], is an instance. Or (Case IV.) the product may be of such a nature that, although the elements of which it is composed and their proportions can be ascertained, the mode of production cannot be discovered from it. In Morgan v. Seaward[8], it was pointed out that the word "use" in the Statute of Monopolies may be read as referring either to the thing produced by the invention, or to the process by which it is produced, and this doubt has not, so far as I know, ever been authoritatively resolved. As at present advised, I am not prepared to accept the rule suggested by Mr. Mitchell as applying to the first class of cases which I have suggested, namely, when the product of the invention is a substance or article already well known. It is clearly applicable to the second class. Whether it applies to the fourth class is, in the state of the authorities, a question of some difficulty. When the invention is in use by other persons than the inventor, in the sense that they use the same process to produce the same result, other considerations arise. Such a case is within the direct prohibition of the Statute of Monopolies, whether the product does or does not disclose the process. The appellant contends that the present case is one of the third class, and that the product of the invention is of such a nature that it discloses the process by which it is produced, and, that, being of such a nature, it was both published to the world by the inventor and used by other persons before the date of the patent.

The case made by the appellant depended upon circumstantial evidence, of which I will briefly state the nature, referring only to facts established by uncontradicted testimony. It was established that the method of illumination by incandescent hoods was well known before the year 1893. The mode of construction of the hoods, namely, by impregnating woven vegetable fibre with solutions of salts of certain rare metals, and then burning off the vegetable foundation, was also well known. Many experiments had been made to discover a metallic basis which would produce a satisfactory result. Up to the year 1893 none of the hoods produced had been a commercial success. In or about the beginning of that year it was discovered that the combination of metallic oxides which is the subject matter of the patent now in question, namely, pure oxide of thorium with a small percentage of 1 to 2 per cent. of cerium, uranium, or prazeodymium, produced very much better results. This was described by one witness as the discovery of the needle in the bundle of hay. Another (Sir James Dewar) said:—"It was really discovering what everybody had a desire, if he possibly could, to discover, and was groping about in the dark to find." A patent for this invention was taken out in England in 1893, but the exact date of the patent does not appear.

The plaintiff company was formed in England under the name of the "Australasian Incandescent Gas Light Company" on 14th March 1893. The memorandum and articles of association were not put in evidence, but it may be inferred from the name of the company that it was formed for the purposes of promoting the business of incandescent gas lighting in Australasia. In June 1893 gas-burners, consisting of incandescent hoods or mantles with the usual appliances, were advertised for sale in Melbourne by the firm of James McEwan & Co. Ltd. under the name of "The Welsbach Gas Burner," and "The New Welsbach Gas Burner," and some of them were sold, and may be taken to have come into use by the purchasers. From September 1893 till the date of the patent, 31st March 1894, and for many years afterwards, the same firm sold gas-burners by the same name as agents for the plaintiffs. It was deposed by the salesman who had charge of that branch of their business during all that period that in May 1893 some hundreds of new burners came into his hands, which were put to use in the warehouse. A large room on the first floor, about 80 feet by 40 in dimensions, was fully lighted with twelve or fourteen of them, and they continued to be so used for many years. They were similar in appearance to those now made by the plaintiffs under their patent, and were sold in similar boxes. The witness said that, so far as he could see, there was no change in the shape of the mantles, or in the brilliancy of the light that they gave, or in the quantity of light which they diffused, before and after March 1894. It appeared from other evidence of expert witnesses that the light which had been given by all the mantles in use before the plaintiffs' invention was far inferior to that given by those made under it. The learned Judge of first instance was of opinion that this witness was thoroughly trustworthy. It is contended for the respondents that, even assuming the accuracy of his evidence, it is not sufficient to establish identity in kind between the mantles sold before and those sold after the grant of the patent. As I have already said, the evidence is circumstantial. In my opinion it is highly improbable that, in the case of a large room entirely lighted by twelve or fourteen hoods, a sudden increase in brilliancy (to the extent of from 50 to 80 per cent.) would not have been noticed by an intelligent observer whose duty it was to push the sale of the article in question. It is also highly improbable that the plaintiff company would have begun their operations in Australasia by sending out obsolete or obsolescent hoods, and when the new discovery was made would not have called public attention to it. Public attention was in fact called in June 1893 to the "New Welsbach Gas Burners," but it was not proved that this was done under the plaintiffs' authority. If, however, still newer and much better hoods were sold after the date of the patent, it would have been to the plaintiffs' interest that the fact should be made known to the public. The plaintiffs' agents, however, whose special duty it would have been to make it public, were not asked to do so. The natural inference of fact is that the new mantles were in use before, as well as after, the date of the grant. If this inference is not in accordance with the truth, nothing would have been easier than for the plaintiffs to prove the actual facts. If the mantles sent by them to their Melbourne agents before April 1894 were of a different kind, and made according to some other specification, that fact is in their own exclusive knowledge. They were directly interrogated on the point, and the gentleman who made the answering affidavit, after saying that inquiries from the plaintiffs' servants in Australia did not enable him to give the desired information, proceeded to object that the defendant was not entitled to an answer to the question "as it would be disclosing the name of the plaintiffs' witness or witnesses." It is suggested that the absence of any evidence is to be accounted for by the fact that the plaintiffs' attention was not directed to the importance of the point, and that they did not anticipate such evidence as that offered by the defendant. The point was however distinctly raised, and no case of surprise is, or could be, made. Upon the evidence I cannot doubt that the conclusion of the learned Judge was correct.

The learned Judges in the Full Court, who were of a different opinion, thought that the only reliable means of ascertaining the composition of one of the mantles would have been to have it analysed by a competent chemist. This would hardly be a practicable mode of proof in a case of a fragile perishable article after the lapse of many years. But there is no reason why circumstantial evidence should not be used in such a case as well as in others. And if the only party in possession of direct evidence withholds it he cannot complain that effect is given to the circumstantial evidence.

Upon these facts it appears that the plaintiffs' invention had been disclosed to the public before the date of the patent so far as such disclosure is to be inferred from the nature of the product of the invention, and also that there was before that time such user of the invention by others than the inventor, (i.e. the persons to whom it was sold), as is to be inferred from the nature of the product. What information then was given by the product itself as to the nature of the invention? It is not disputed that the relative proportions of thorium and the other metallic base, whatever it was, could be discovered by analysis. The method of applying the metallic base to the vegetable fibre in the form of a solution of nitrates was disclosed by the specification for patent 4472, granted in 1886, and it appeared from the evidence of the expert witnesses, who spoke of the solution as "the lighting fluid," to be a matter of common knowledge. It is clear, then, that the product of the invention was of such a nature that anyone conversant with the subject and applying the common knowledge of the time could, with the aid of the information given by the product itself, have brought about the same result. It was, however, contended by Mr. Coldham that this is not sufficient, and that the validity of the patent is not affected unless the Court is able to draw the further inference of fact that some person actually made the necessary analysis and so discovered the nature of the invention. In the case of In re Miller's Patent[9], the contrary proposition was treated as too clear for argument. In my opinion the rule laid down in In re Gaulard & Gibbs' Patent[10], as to prior publication by writings, is equally applicable to publication by putting the thing itself in public use when the thing is of such a nature as to disclose its composition and the mode of making it, and the Court is bound to draw the inference that the information actually available to the public is in their possession. I am, therefore, of opinion that the invention had been published to the world in Victoria before the date of the patent. The product of this particular invention is, I think, to be regarded as a new thing. And, since the manner of producing it was disclosed by the thing itself, the invention was, in my opinion, also used by others than the inventor within the meaning of the Statute.

In this case only three issues were contested before us, that is to say:—that the alleged invention is not new; that it had been published in Victoria prior to the application for the patent; and that it was publicly used in Victoria prior to the application for the patent.

The company was incorporated on the 14th March 1893, and the specification of the patent taken out on the 31st March 1894, No. 11,247, is as follows:—

With regard to the plaintiffs' mantle it was not denied in argument that its constituents and true proportions could be ascertained by analysis; but it was said that was not enough, because the main portion of the process, that is to say the part played by the aqueous solution of the nitrates, was still a matter to be discovered. Substantially, at any rate, it did not remain to be discovered, because, by a patent, No. 4472, issued on the 5th February 1886 to the assignor of the plaintiff company, I find that, in a part of the process, the patentee claimed the use of a fine fabric, preferably of cotton previously cleansed by washing with hydrochloric acid, which fabric is saturated with an aqueous solution of nitrate or acetate of the oxides of the rare metals, and so forth. As to the facts which came out in evidence before Hodges J., there were sales by James McEwan & Co., first apparently on their own account, and secondly as agents for the respondents; (a) on their own account for some time in 1893, not later than June, and (b) as agents for the respondents from September 1893 till long after the issue of the patent. I shall not follow His Honor through the details of the evidence as to the identity in kind of the mantles sold by James McEwan & Co., whether independently or as agents before the patent was issued, and those sold by the company afterwards as agents under the patent. Evidence was given by a witness who was in a responsible position in McEwan & Co.'s establishment, who had to look after this branch of their business, and who, more than anyone else, probably, would be likely to notice any sensible difference in appearance in the shape, make or luminousness, or in the colour of the light which the mantles gave. It would be most strange that there should be any marked difference in any of those particulars which would not be observed by him. At any rate he gives evidence that, during the period anterior to the agency of McEwan & Co. and from that onwards and during the latter period after the issue of the patent, he could observe no difference, in any of those particulars, in the appearance of the mantles. That seems to me to be primâ facie evidence. Unanswered it is strong evidence. Of course I do not forget that, when novelty is in issue, the presumption is that the invention is novel once the patent is produced. The case of Amory v. Brown[11] cited by Mr. Coldham, seems to establish that proposition, and Lindley L.J. in Harris v. Rothwell[12], cites it with approval as settling the question. When the question is one of novelty the plaintiff establishes a primâ facie case by proving his patent, so that the burden is then thrown on the defendant of proving prior use or prior publication, but when he has given express, even though not overwhelming, evidence to rebut the tacit presumption, does it not become fatally strong when no attempt is made to deny or in any way meet it? It is not disputed by the respondents that all the mantles sold from September 1893 were Welsbach mantles. Were those sold before that time different? Evidence having been given that they were the same in appearance, that is, in shape, make, luminousness, colour, &c.—in fact that McEwan & Co.'s employé could not find any difference—it would have been interesting to know whether the mantles previously on sale were withdrawn in September when the agency was conferred, and replaced by fresh ones; or whether the previous stock was allowed to run down to nothing and new stock put in its place in September; and similarly as to the period between September 1893 and the issue of the patent. The defendant having elicited primâ facie evidence that the mantles sold during the three periods were precisely alike—indeed identical in effect and so presumably in composition—the fact that no attempt is made to shake the evidence in its main features or to countervail it must have been most impressive to the learned Judge who tried the case, as it would have been to any of us. The expert evidence has been sufficiently dealt with by His Honor the Chief Justice, and I need only state that I have come to the same matter of fact conclusion as he has on that question.

Now the Statute of Monopolies (21 Jac. I. c. 3), after prohibiting monopolies, enacts in sec. 6 that: "Any declaration before mentioned" (against monopolies) "shall not extend to any letters patents and grants of privilege for the term of fourteen years or under, hereafter to be made, of the sole working or making of any manner of new manufactures within this realm, to the true and first inventor and inventors of such manufactures, which others at the time of making such letters patents and grants shall not use." The Victorian Patents Act 1890, defines "invention" as "any manner of new manufacture which might be in England and Wales the subject of letters patent and grant of privilege within section six of the Statute of Monopolies," &c. So that the definition clause of the Victorian Patents Act 1890 brings into use, for the purpose of defining "invention," the very phraseology of the Statute of Monopolies. Now the question here is really in a broad way one of novelty. There were some expressions of Mr. Justice Story in the case of Pennock & Sellers v. Dialogue[13], which are worth citing. He said "It has not been, and indeed cannot be denied, that an inventor may abandon his invention, and surrender or dedicate it to the public. This inchoate right, thus once gone, cannot afterwards be resumed at his pleasure; for, where gifts are once made to the public in this way, they become absolute." In illustration of that he cited the case of dedication of a highway. The following passage from a well known text book was cited with approval by high authority in the House of Lords in 1877 on a question of prior publication. "If the public once becomes possessed of an invention by any means whatever, no subsequent patent for it can be granted either to the true and first inventor himself, or any other person, for the public cannot be deprived of the right to use the invention, and a patentee of the invention could not give any consideration to the public for the grant, the public already possessing everything that he could give." That passage is from Hindmarch on Patents, 1st ed. (1846), p. 33, and Lord Blackburn says in Patterson v. Gas Light and Coke Co.[14]:—"This, is, in my opinion, a correct statement of the law." The passage is cited again with approval by Fry L.J., in Humpherson v. Syer[15], and as Lord Blackburn says in the case I have mentioned, the consideration for a patent is the communication to the public of a process that is new. It may be deprived of novelty in many ways, but in three principal ways:—(1) by prior publication, i.e. by publication of a sufficient description which would enable people to construct or make the thing in question by a specification which would give similar light and information, and published before the patent; or (2) by prior user, i.e., user before the patent, except for the purposes of mere experiment and under confidence, by the inventor himself; or, (3) by user by the public in the ordinary and open way, as, for instance, by buying it in the open market and using it for the purpose for which it is designed. So, as to the issue of "true and first inventor," that in most cases is a mere question of novelty and generally depends upon the words which are to be found in the Statute of Monopolies. In that connection I refer to the judgment of Lindley L.J., in which Cotton L.J. concurred, in Harris v. Rothwell[16], which I think will be found to be enough to show that the issue of "true and first inventor" should be found for the defendant in an action for infringement, if the jury, where there is one, or the Judge without a jury, thinks that the invention claimed by the plaintiff has "been so published in this country as to have become known to anyone here." There cannot be any distinction in reason between these two cases; on the one hand, the deposit in an accessible place in a public library of a book or scientific journal containing a description of the process; and on the other hand, a number of sales of the product, where that is of such a nature as to give the means of knowledge of the process to those who are conversant with the subject. And in the case of Croysdale v. Fisher[17] Pollock B. puts the matter very pointedly when he says:—"Whether a patent was anticipated by a previous specification or by previous publication in a book, or by previous user, the principle is the same, because in each case it must be shown, in order to effect a good anticipation, that some person before the date of the patent has communicated to the public that which the patentee now claims." Here, of course, the question is whether that which the plaintiffs now claim has been communicated to the public, either in writing or in any other way, because writing is not essential; whether it has been published to the world by the user of the plaintiffs, or by the user of anyone else, or by a simple divulgation. The question as to publication may be put in the following words used by Fry L.J. in Humpherson v. Syer[18]:—"Is it the fair conclusion from the evidence that some English people, under no obligation to secrecy arising from confidence or good faith towards the patentee, knew of the invention at the date when the plaintiff took out his patent?" If, under no such seal of secrecy, people did know of it, and knew of it in such a way that their knowledge was effective, then the inventor cannot subsequently take out a patent for an invention of which knowledge has been so gained; or, if he does take it out, that patent is void. And in that case His Lordship held that the disclosure of the gist of the plaintiff's invention by a third person —the defendant—to a person who rested under no obligation of confidence was enough to constitute publication. Hoe & Co. v. Foster & Sons[19] was a case in which there was a user of printing machines by the patentees and the newspaper owners in conjunction, for a fortnight up to the date of the grant. The machines which were imported by the patentees for the newspaper owners under contract, made several trial runs, and some small defects were remedied. On all occasions the papers were sold as printed by the machines, which were running perfectly a week before the patent issued. The machines were seen by several persons, and though it was alleged that the user was under a pledge of confidence, that excuse was brushed aside, and it was held by the Court of Appeal that the user described was fatal to the patent. Vaughan Williams L.J. concluded his judgment thus[20]:—"I am aware that the Statute of Monopolies speaks of user by others, and at the same time there is no doubt that a user by the patentee does prevent his subsequently taking out letters patent, and it seems to me that really, where-ever you have a patentee supplying the patented article in such a way and to such an extent that he cannot recall the patented article—that he cannot impose any obligation of secrecy—immediately you have that state of things, it becomes impossible for the inventor subsequently to take out letters patent." Of course, in that case the thing patented was a machine, the improvements in which, the subject of the patent, would be disclosed to a competent person by inspection. But the passage shows clearly the learned Lord Justice's opinion that the avoidance of a patent by user on the part of the inventor himself stands on the ground of publication or gift to the public—in effect the ground of want of novelty; while the user by others is referable to the explicit exception in the Statute of Monopolies. Both the other members of the Court—Lindley M.R. and Chitty L.J.—based on this user their decision on the two issues—first, that the invention was not new, because the patentees themselves had used it before grant and so had published it, and, secondly, that it had been used by others before the date of the patent. As Moulton, then Q.C. now L.J., said in the course of the argument[21]:—"The doctrine of prior publication rests on the word new, that of prior user rests partly on the same word and partly on the words which others at that time of making such letters patents and grants shall not use." The respondents relied a good deal on the case of In re Miller's Patent[22]. In that case Romer J. had pronounced against the validity of a patent on the ground that there had been prior user, inasmuch as the product of the invention had been made and sold in England by the inventors before the date of their patent. In coming to that conclusion he had found that, in the state of chemical knowledge of the time, the entire composition of the product as to ingredients and proportions could have been discovered by analysis, so that the prior user amounted to publication. Then he said this[23]:—"The findings I have made render it unnecessary for me to decide the point of law which would otherwise have arisen, which I think is one of considerable importance, and one upon which I should have desired further consideration but for the findings of fact I have made. Even assuming that the Magnolia Metal sold could be said to have been a metal made according to a secret process which could not have been detected in the metal by analysis, it would be still a question whether the true owner of this patent could, under the circumstances, have said that that sale would not invalidate the patent. It is a point which one day will have to be determined, and a point upon which I need at present express no final opinion, because, although it has to be considered, I think it might be said to be still open in some respects for further consideration by the Court, whether a person who discovers a secret process and uses the benefit of that secret process to manufacture an article which he sells publicly and commercially, and obtains considerable benefit from it, and then seeks to obtain a patent for that very same process, is entitled to hold that patent under these circumstances. That is a question which I think will require one of these days to be carefully considered." Without pre-judging that question, which, in the view I take of the case, does not arise here, I should infer from the frame and tone of His Lordship's remarks that he inclined to the opinion that even under those circumstances the patent could not hold. In the present case we might have arrived at the consideration of the point indicated by Romer J., but for the fact that in the specification of 1886, No. 4472, the assignor of the respondent company had given to the world the knowledge that a part of the process by which a mantle of the rarer earths could be made was the saturating of some light fabric such as cotton with an aqueous solution of the nitrate or acetate of the oxides of such earths. That knowledge may be taken together with the clear evidence, not seriously disputed by the respondents' counsel, that the components of the mantle in the 1894 patent No. 11247 could be detected by analysis. With this sum of knowledge the proved sales of the mantles by or for the patentees before the issue of the patent now in question were incontestably a publication. As the question suggested by Romer J. does not arise in this case because of the last mentioned facts, it is clear to my mind that in this case both publication, in the sense of their own prior user by the patentees, has been proved, and also a user on the part of the public by buying the product which amounts to such a user within the Statute of Monopolies as would of itself defeat the patent. In respect of that user, I may conclude with some words that were used by M'Lean J. in delivering the judgment of the United States Supreme Court including Marshall C.J. and Story J. in the case of Shaw v. Cooper[24]:—"Whatever may be the intention of the inventor, if he suffers his invention to go into public use, through any means whatsoever, without an immediate assertion of his right, he is not entitled to a patent; nor will a patent, obtained under such circumstances, protect his right."

Loath as I am to be a party to the avoiding of a patent which has apparently been gained by a bonâ fide inventor, still the objections which have been taken by the appellant in this case are too well founded to be overruled, and the appellant has, in my view, abundantly sustained the onus cast upon him in regard to each of the three issues I have mentioned. I am, therefore, of opinion that this appeal must be allowed.

Although the argument in this case covered a very wide field, the questions for decision are within a small compass. If the respondent company's patent was valid there has clearly been an infringement. The validity of the patent, however, was impeached on several grounds, two of which only are mentioned on this appeal, namely, that the invention was not new, and that it was publicly used in Victoria prior to the application for the patent. In support of both grounds the appellant relies upon the same state of facts, that before the application James McEwan & Co. were selling as the respondents' agents gas-mantles manufactured by the respondents of the same materials in the same proportion and by the same process as described in the company's patent, and that there had been, before the application for the patent, a public use in Victoria of the mantles so sold both by the persons who bought them from McEwan & Co., and by McEwan & Co. themselves, during the period when they were not the company's agents. A public use in Melbourne by the Metropolitan Gas Co. was also relied on, but, as I am not satisfied of the identity of the mantles used by the gas company with those made under the patent, I leave that issue of fact out of consideration.

It was admitted by the appellant that the onus of these defences was upon him. Hodges J. found that the appellant had established them, and entered judgment for him. The Supreme Court, on appeal, reversed that judgment on the facts, not questioning the propositions of law upon which Mr. Mitchell, counsel for the appellant, relied. Although other matters were dealt with in the judgment, the main reason for the decision is contained in the following passage[25]:—

The necessity for establishing the identity mentioned cannot be disputed, and if the Supreme Court were right in their decision on the facts, the defence must fail. On the other hand, if the judgment of Hodges J. on the facts is right, the further question arises whether, upon the law applicable to the facts as found by him, the respondents are entitled to succeed. In my opinion the decision of Hodges J. on the facts was right, and I concur in the reasoning by which my learned brother the Chief Justice has arrived at the same conclusion. I wish, however, to add something as to the mode and sufficiency of proof in such cases.

The onus of establishing the identity of the mantles sold by McEwan & Co. before the application for the Victorian patent with those sold afterwards was on the appellant. But it was not necessary to establish that identity by the most complete and exhaustive proof possible. No doubt, the most satisfactory evidence would have been that of experts based on chemical analysis of the mantles in use before the patent. But the appellant was not tied down to that mode of proof. Proof by admission or by circumstantial evidence was equally admissible, and might be equally cogent. If he established in evidence circumstances from which the only fair and reasonable inference, in the absence of explanation by the respondent company, was the identity of the mantles, that is all he was bound to do in discharge of the onus which the law imposed on him in the first instance. In such cases there is a shifting onus as is clearly explained by Sir James Stephen in his Digest of the Law of Evidence, 5th ed., at p. 111, in the following passage:—

In applying this rule each case will depend upon its own circumstances. Mere proof of similarity of appearance in form, colour, intensity, and power, of the light, would not of itself have been sufficient. Regard must be had to the skill, knowledge, and opportunities of forming a sound judgment, possessed by the witnesses. But in this case I have no hesitation in holding, in view of Mr. Lacey's experience and knowledge as a salesman of these mantles, and his opportunities of drawing a sound conclusion, that Hodges J. was right in deciding that the onus of explaining the appellant's primâ facie case was thrown on the plaintiff company, and that, in the absence of such evidence, the identity relied on by the defendant must be taken as established. That being so, it becomes essential to consider the questions of law argued before us with much research and ability on both sides. I do not consider it necessary after the full exposition of my learned colleagues to discuss these questions in detail. I shall confine myself to stating what I conceive to be the law applicable to the facts as found by Hodges J. The defences of want of novelty and prior user in this case must be kept separate. They both arise under the sixth section of the Statute of Monopolies, but are founded on different portions of its provisions. I propose to deal only with one of the defences, "want of novelty," that to which the argument was principally directed. An invention may consist in the new or improved process of manufacture of a known article, or it may consist in the manufacture of some new article or compound. Where the process of manufacture is used by the inventor openly and publicly in the way of business within the realm, that is to say, in the country in which he afterwards makes his application, his patent will be void whether the product does or does not in itself afford the public a knowledge of the invention. But where the process of manufacture is not publicly used by the inventor in the country in which his application is made, but the product has been sold there publicly before his application, then the question whether the product does or does not afford a knowledge of the invention becomes of vital importance.

Frost in his work on Patents, 3rd ed., at page 137, deals with a position of the same kind arising in England. "A question," he says, "closely connected with the above is, will the publication in this country of an article made abroad be a publication of the invention by the exercise of which the article was made? If the article made abroad shows on the face of it how it was made, its sale in this country is a publication of the invention; but there is authority for saying that such will not be the case if the article does not show on the face of it how it was made."

The same principle must apply to a product of an English invention made in England and imported into Victoria, for it must be taken in this case that the mantles were made in England and sold in Victoria in the ordinary way of business before the patent in Victoria was applied for. The law is now too well established to be disputed that an inventor may by public sale of the product of his invention, if the nature of the invention can be ascertained by the public from an examination of the article sold, disentitle himself from afterwards patenting the invention. The cases from Wood v. Zimmer[26] down to Hoe & Co. v. Foster & Sons[27] in laying down and acting upon that rule have assigned many reasons, some good and some very unsubstantial, for its existence. Perhaps the most practical is that stated by Vaughan Williams L.J. in his judgment in the latter case as follows[28]:—

"I am aware that the Statute of Monopolies speaks of user by others, and at the same time there is no doubt that the user by the patentee does prevent his subsequently taking out letters patent, and it seems to me that really, wherever you have a patentee supplying the patented article in such a way and to such an extent that he cannot recall the patented article—that he cannot impose any obligation of secrecy—immediately you have that state of things, it becomes impossible for the inventor subsequently to take out letters patent." In other words, when he has sent the product carrying the secret of the invention on its face into the world for the purposes of profit he cannot be heard any longer to say that his invention is new. I cannot assent to the proposition in the broad terms contended for by Mr. Mitchell that, where there has been a sale by an inventor in the ordinary way of business of the product of his invention, he loses his right afterwards to apply for a patent whether the product does or does not disclose in itself the secret of the invention. None of the cases when examined authorize such a statement of the law. In all of them, where the sale of the product was relied on, the product in itself disclosed the invention. The most important question in this case therefore is, did the mantle contain within itself information accessible to the public of the nature of the invention? In dealing with that question the common knowledge existing at the time as to the subject matter of the invention must be considered. As to what is common knowledge the exposition of Jessel M.R. in Plimpton v. Malcolmson[29] is worth quoting:—

"When," he says, "you say a thing is known to the public and part of common knowledge, of course you do not mean that every individual member of the public knows it. That would be absurd. What is meant is that if it is a manufacture connected with a particular trade, the people in the trade shall know something about it; if it is a thing connected with a chemical invention, people conversant with chemistry shall know something about it. And it need not go so far as that. You need not show that the bulk, or even a larger number, of those people know it. If a sufficient number know it, or if the communication is such that a sufficient number may be presumed, or assumed, to know it, that will do. Now how are they to know it? They are to know it by being told of it, or informed of it in some way. You may show that they know it, by showing that the trade had commonly used it. That is the best evidence you can have. You may show the thing was known because it was used and brought into practice, which is a case I have not now to consider. But you may show they knew it in another way—that it was published, or made known to the public. I use the word published in that sense. How made known to the public? It has been held that if it is in a specification, certainly in a modern specification, which has been enrolled in the Patent Office, and not published besides, that will do. And it has also been held that, as a common rule, if the description has been printed in England, and published in England, in a book which circulates in England, that will do. But, after all, it is a question of fact. The Judge must decide, from the evidence brought before him, whether it has in fact been sufficiently published to come within the definition of being made known within the realm."

In this case the evidence clearly establishes that everything necessary for the making of the patented article, except the nature and proportions of the constituents forming what has been spoken of as the "lighting fluid," was common knowledge in the sense so explained by Jessel M.R. at the time when these mantles were being sold in Melbourne before the application for the patent. It was the proportion in which cerium was mixed with thorium in the solution to be applied to the hoods which constituted the invention. The Victorian patent No. 4472 described the method of preparing the hood and applying the metallic base. If the constituents and proportion of the "lighting fluid" in which the hoods were to be saturated could be discovered by analysis the invention could be ascertained. There was no difficulty in discovering by analysis the constituents of the solution or mixture to be applied to the mantle or the proportions in which those constituents were used. In fact, part of the evidence given by the respondent company to establish infringement was founded on the analysis of the appellant's mantles. There was clearly no more difficulty in the discovering of the invention from the product in this case than there was in the case of the "Magnolia Metal," the subject of inquiry in the case In re Miller's Patent[30]. The observations of Romer J. in that case are very applicable to the supposed difficulties of the analysis in this:—"But it was said, and, as I have stated, that is the point now before me, that at the time the metal was so sold chemists analysing it would not have been able to find out the constituents—at any rate, so far as concerned the bismuth. That is the point I have to try. Now, I am satisfied that the respondents have failed on that. The bismuth could have been detected, and I am satisfied that, if reasonable care and reasonable skill were used by analysts before the date of this patent, when this Magnolia Metal was sold, its constituents and the proper proportion of those constituents could have been ascertained, so that, the metal being sold, its constituents and their proportions could have been known, and it could have been made. There was no difficulty, I wish to point out, and no secret or invention in making the alloy, when you once knew the constituents and the proportions of those constituents... So that you have a public sale of a metal, as I am pointing out, not of a metal made by any secret process at all, but a public sale of a metal which if anyone wanted to find out what it was made of and how it was made, he could have done so without difficulty by analysis."

Mr. Coldham strenuously urged that the particular method of applying in their proper proportions the metals forming the "lighting solution" in the form of nitrates was not discoverable by analysis. But the answer to that contention is that sufficient information to give that knowledge was accessible to any expert who examined the former patent No. 4472 already mentioned.

Under the circumstances, I am of opinion that Hodges J. came to a right conclusion in finding, as he must be taken to have found, that to any person skilled in the chemistry of incandescent lighting, and having the knowledge then accessible in Victoria to any such person, the nature of the respondents' patent was ascertainable from an examination and analyis of the mantle itself. Such being the case the sale in the ordinary way of business of these mantles in Victoria before the patent was applied for made it impossible for the respondent company to claim that the invention was new within the meaning of the Statute of Monopolies. On this ground the appellant is entitled to succeed. I therefore agree that the appeal must be allowed and the judgment of Hodges J. restored.

The issues raised in this cause have been numerous, and on nearly all there has been agreement before the primary Judge and the Full Court; but the difference of opinion on one point involves the success or the failure of the plaintiffs. The point is, has the defendant adduced sufficient evidence to satisfy the Court—in the absence of counter evidence—that the patented manufacture was used in Victoria by others than the plaintiffs at the time of the grant of the patent.

Hodges J., who conducted the trial, said yes: the Full Court said no. Having had the advantage of hearing counsel's criticisms of both the judgments, I have come to the conclusion, not only that there was no sufficient ground for rejecting the finding of Hodges J., but that that finding was right.

It seems that the patent in question is the third of a series of patents under the Welsbach system. In the Victorian patent 4472 of 1886, there is found the use of certain rare oxides. In the British patent 3592 of 1886, there is found the use of pure thorium oxide and certain other alternative materials. But mantles under these patents did not prove very successful; and between 1890 and 1893 the patentee discovered that the oxide of thorium, pure, gave no light. In 1891 or 1892, mantles in Europe began to show a great increase in illuminating power; and the new process was represented in the Victorian patent in question. The plaintiffs obtained their patent for the new process in England at some time in 1893, and in Victoria on the 31st March 1894. Now, it seems that about May 1893, certain incandescent mantles came from England to McEwan & Co., of Melbourne. It is conceded that they came from the plaintiff company, which was incorporated on the 14th March 1893, and was at first called The Australasian Incandescent Gas Light Co. Ltd.; and they were in boxes similar to those in which the plaintiffs' mantles made according to their patent now come. These were placarded for sale in McEwan's windows. Some were burning in the windows and in the warehouse during part of every day for months, and hundreds were sold during 1893 and 1894. According to their answers to interrogatories, the plaintiff company believes that McEwan & Co. were its agents in respect of these mantles from September 1893, selling them under the name of the "Welsbach gas-burner," and the "New Welsbach gas-burner." According to the evidence of Lacey, a salesman of McEwan & Co., the illuminating power was as great as that of the Welsbach burner of to-day, which are made under this patent. Now, the plaintiffs have only taken out two patents in Victoria—No. 4472 of 1886, and this patent 11,247 of 1894. The patent of 1886 was for a hood made of fabric impregnated with a compound of oxide of lanthanum and zirconium or of these with oxide of yttrium; and it is admitted that a hood made under this patent would not give any such illumination as the present Welsbach hoods. Hoods made in accordance with the patent of 1894 would give, and do give, this illumination. It is said, however, that the plaintiffs may have sent out to McEwan & Co. in 1893 hoods made, not in accordance with the important discovery of 1891-2, but in accordance with the British patent 3592 of 1886. This is the pure thorium oxide patent. It is sufficient to say that the hoods made under this patent were a failure as to illuminating power; and there is no evidence that they ever reached Melbourne. On such evidence, it is, to my mind, plain, that in the absence of evidence to contradict or to explain it, the only inference to be drawn is that the hoods of the plaintiff company sold in 1893 and 1894, having the same illuminating power as the plaintiffs' present day hoods, were made in accordance with the same process as the present day hoods. If this inference is wrong, it was for the plaintiffs to contradict it by evidence. The plaintiffs had ample warning and ample opportunity, as shown by the dates of the defendant's objections, and of the commission to London, and as shown by the Judge's notes of the trial. If the mantles sent from England by the plaintiffs in 1893 were not made in accordance with the present day mantles, that fact would be peculiarly within the knowledge of the plaintiffs and their servants; and they have not adduced any evidence whatever on the subject. Probably the course taken by the plaintiffs' advisers was taken in the exercise of a wise discretion; and it has actually led to success in the Full Court. But I agree with Hodges J. I also think that when the primary Judge has come to a definite finding on a question of fact, as to which there is evidence, it is not the function of an Appeal Court to determine what conclusion its members would have arrived at if they were Judges of first instance. It is their duty to dismiss the appeal unless the appellant satisfy them that the primary Judge has been clearly wrong: Allen v. Quebec Warehouse Co.[31]; Colonial Securities Trust Co. v. Massey[32]. I do not refer merely to cases which turn on the credibility or veracity of witnesses. The mode in which witnesses are cross-examined, the kind of questions put, and their mode of answering, as well as the whole conduct of the trial, are matters which the primary Judge only can satisfactorily weigh. I refer to this point particularly, because it has been urged that Lacey's evidence was weakened by his cross-examination. It is quite possible that the answers, as printed, are to be explained by the form of the leading questions put by counsel in due exercise of his rights as cross-examiner.

But the plaintiffs' counsel have also, in the argument before us, boldly impugned the accuracy of the law both of the primary Judge and of the Full Court. They insist that it is not enough for the defendant to show prior user of the invented article by others—that there must be such prior user as puts the public in actual possession of the invention—such user as supplies the public, not only with the means of knowledge, but with actual knowledge how to reproduce the article. The authorities cited have been so fully discussed in the judgments of my colleagues that I need say little. Personally, I have not felt much doubt on the point from the first, although I have found difficulty in tracing the precise application of some of the phrases used in the numerous judgments. Sec. 6 of the Statutes of Monopolies (21 Jac. 1 c. 3) is the fountain head of our patent law. It is incorporated by the Victorian law by the definition of in sec. 3 of the Patents Act 1890. The only letters patent of monopoly permitted by law are those for "the sole working or making of any manner of new manufactures within this realm, to the true and first inventor and inventors of such manufactures, which others at the time of making such letters patents and grants shall not use." I feel strongly that in cases turning on Statutes, however old, the more steadfastly we keep our eyes on the words of the Statutes the better. The Statute is the beacon light for our guidance; and we must not let ourselves be distracted by the multitude of cases, as by the lights of mere lamps along the shore. Of course, if the Statute has during a long course of years received one consistent interpretation, we should accept that interpretation; but otherwise we should not depart from the natural, reasonable meaning of the words. The decisions in the United States, in particular, have to be used with great reserve and caution, inasmuch as the words of the United States Statutes from 1793 onwards, are very different from the words of the Statute of Monopolies (see Bates v. Coe[33]. In the present case, McEwan & Co. publicly used these mantles before they became agents as well as after. They sold the mantles in hundreds, to various customers; and, presumably, the customers used what they bought. There is nothing in the words of the Statute about knowledge on the part of those shown—nothing about degree of disclosure of the mode of manufacture. No doubt, expressions may be found, in many of the cases, used by Judges who have not thought it necessary always to point out the distinction between user by the patentee himself, and user by others. Where it is alleged that a patentee has, before his patent, given the world the benefit of the invention, thrown it out freely to the world, as it were, the degree of disclosure, the amount of knowledge communicated, may become very material. Moreover, the defences of "not true and first inventor," "not new" (or "known before"), "prior publication," "anticipation," &c., rest on facts very similar to "prior user," facts which sometimes merge almost insensibly in the facts on which "prior user" is based. But I cannot find in the authorities cited any decision to the effect that, if an invention has been used before the patent by others than the patentee, the defence is bad unless it be also shown that the public was supplied with actual knowledge, or potential knowlege, how to reproduce the invention. The case of Gill v. Coutts & Sons & Cutler[34] goes a long way towards bringing the reasoning back to the words of the Statute. The present case, however, is comparatively simple. It is the case of an invention—not of a new process, but of a hood having certain constituents—a hood which is an improvement on a previous hood made with other constituents; and, assuming that such a finding is necessary, Hodges J. has found that "the article in itself presented such means of knowledge to the public as to enable anyone of ordinary competence to reproduce the article." I accept this finding as correct, and I regard it as sufficient to justify the judgment for the defendant, if knowledge or means of knowledge should be treated as an element in the defence of prior user.

[1] Holt N.P., 58.

[2] 39 Newton's Lond. Journ., 158.

[3] [1837] EngR 176; 2 M. & W., 544; 1 Web. Pat. Cas., 167.

[4] [1860] EngR 746; 2 B. & S., 194, at p. 226, per Pollock C.B.

[5] 4 R.P.C., 407.

[6] 1 Web. Pat. Cas., 530.

[7] 15 R.P.C., 205, at p. 212.

[8] [1837] EngR 176; 2 M. & W., 544.

[9] 15 R.P.C., 205.

[10] 7 R.P.C., 367.

[11] L.R. 8 Eq., 663.

[12] 4 R.P.C., 225, at p. 229.

[13] [1888] USSC 161; 2 Peters, 1, at p. 15.

[14] 3 App. Cas., 239, at p. 244.

[15] 4 R.P.C., 407, at p. 414.

[16] 4 R.P.C., 225, at p. 230.

[17] 1 R.P.C., 17, at p. 21.

[18] 4 R.P.C., 407, at p. 414.

[19] 16 R.P.C., 33.

[20] 16 R.P.C., 33, at p. 40.

[21] 16 R.P.C., 33, at p. 36.

[22] 15 R.P.C., 205.

[23] 15 R.P.C., 205, at p. 212.

[24] [1833] USSC 19; 7 Peters, 292, at p. 322.

[25] (1906) V.L.R., 677, at p. 685; 28 A.L.T., 155, at p. 157.

[26] Holt N.P., 58.

[27] 16 R.P.C., 33.

[28] 16 R.P.C., 33, at p. 40.

[29] 3 Ch. D., 531, at p. 556.

[30] 15 R.P.C., 205, at p. 211.

[31] 12 App. Cas., 101.

[32] (1896) 1 Q.B., 38.

[33] 8 Otto, 31.

[34] 13 R.P.C., 125.

High Court of Australia

Cullen v Welsbach Light Company of Australasia Ltd [1907] HCA 3; (1907) 4 CLR 990 (18 March 1907)

HIGH COURT OF AUSTRALIA