Filta-Matix (Pty) Ltd v Freudenberg and Others (258/1996) [1997] ZASCA 110 (27 November 1997)

2

HARMS JA:

This appeal concerns the validity and infringement of patent 76/5566. The patent, entitled "A Gas Filter Element", belongs to the first respondent, and the second and third respondents are registered licensees under the patent by virtue of which they could join as plaintiffs in an infringement action. The appellant, the defendant before the Court of the Commissioner of Patents (MacArthur J), admitted that it had copied the commercial product made and sold by the plaintiffs, but denied infringement of the patent. In addition, it counterclaimed for revocation of the patent on several grounds of alleged invalidity. MacArthur J found in favour of the plaintiffs on all aspects of the case, issued an interdict and ordered an enquiry into the damages suffered. He granted leave to appeal to this Court. For the sake of convenience I shall refer to the respondents simply as "the patentee".

The patent was granted pursuant to a convention

3

application first filed in Germany on 17 September 1975. That is the effective date of the patent - the date on which its validity has to be decided. The initial term of the patent was sixteen years, which was extended for a further five years. The validity of the patent is to be judged in terms of the repealed Patents Act 37 of 1952. (See s 3(1) of the Patents Act 57 of 1978.) Judgment was given in the court below on 1 2 March 1996 and the patent lapsed on 16 September 1996.

The specification:

What follows is a discussion of the more pertinent statements in the specification.

Gas filter elements are employed for separating particles (such as dust) that are in suspension in a gas (for instance air) from the gas in, e g, ventilating systems.

High surface filter units consisting of sets of filters of mats of fleece material installed in rigid supporting screen structures were known. The mats were in a

4

V-shape, presumably to increase the surface area of the filter because the greater the surface area the better the filtering properties of the element. These filters were highly effective but had several disadvantages, being uneconomical, difficult to replace and giving problems in controlling the dust proof connection.

Units similar in shape but without rigid supporting screens were also known though not widely used because of other inadequacies. Their filter pockets were made from glass fibre or textile fibre and sewed, glued or spot welded. These operations are labour intensive and expensive. The individual filter pockets bulge under operational conditions, resulting in uneven flow conditions and allowing the pockets to touch each other, thus impairing their filtering efficiency. Fluttering motions occur causing unsatisfactory performance and damage to the pockets because of the loosening of fibres. Spacers were used to prevent bulging but the known method of fixing these was time-consuming and caused a reduction of the

5

active filter surface.

Against this background the specification sets out the object of the invention in what can be called the promissory clause:

"The object of this invention is to provide a pocket filter element which attains the performances of rigid high surface filters as regards the degree of dust separation and dust storage capability while avoiding the aforementioned disadvantages."

The so-called consistory clause follows:

"This object is achieved according to the invention by a gas filter element comprising a supporting frame and a plurality of wedge-shaped filter pockets undetachably connected with this frame, and characterised in that the pockets are of a suitable filtering medium and have self-supporting properties provided by a continuous, edge-trimmed weld and spacer elements secured to the filter medium in the air flow direction and, if desired, by additional welded stiffening zones, and that the edges approached by the air flow are continuously and undetachably connected with the supporting frame which is made of a hard foamed material and is connected directly with the filtering medium by a foaming process."

The specification explains the invention with

6

reference to two drawings and concludes with a more specific promise:

"The advantages achieved by the filter element according to the invention reside especially in the fact that by its application the filter can be rapidly exchanged, even by untrained labour, while ensuring that break through places of dust are avoided. The filter pockets do not tend to flutter and the filter element achieves the performance of rigid high surface filters as regards its dust separation and storage."

There are five claims. Claim 1 conforms with the consistory clause. In the quotation that follows the numerals in brackets (according to the German tradition of claim drafting) are references to the drawings . The enumeration by way of letters represents the division of the claim into integers proposed by the parties and adopted by the court a quo. The claim reads:

"1. (a) A gas filter element

2. comprising a supporting frame (2) and

3. a plurality of wedge-shaped filter pockets (1)

4. undetachably connected with the frame, characterized in that

7

5. the pockets are of a suitable filtering medium and

5. have self-supporting properties

5. provided by a continuous edge trimmed weld (3) and

(h) spacer elements (4,6)

(i) secured to the filter medium in the air flow

direction, (j) and, if desired, by additional welded

stiffening zones (7), (k) and that the edges approached by the air flow

are continuously and undetachably connected

with the supporting frame (1) which is made of a hard foamed material (m) and is connected directly with the filtering

medium by a foaming process."

I do not propose to deal with claims 2,3 and 4 because they are dependent upon and narrower than claim 1 and therefore do not contribute to the resolution of any issue on appeal. Claim 5 should be mentioned. It is a so-called omnibus claim and claims -

"5. A gas filter element, substantially as described herein with reference to the accompanying drawings."

The description of the invention with reference to

8

the drawings differs from the description of the invention in the consistory clause but these differences need not be dealt with at this stage of the judgment.

Reverting to claim 1, the meaning of "self-supporting" was the subject of much evidence and debate because of its importance for the decision on both infringement and invalidity. It is an ordinary English word with no special or technical meaning. Evidence, expert or otherwise, is impermissible to explain it. According to the Shorter Oxford English Dictionary it means "(of a physical object) not requiring the usual support". "Support" has many shades of meaning but in the context of the invention the following are apposite: "1. To bear the weight of, especially from below. 2. To hold in position; prevent from falling, sinking or slipping. 3. To be capable of bearing; withstand." (American Heritage Dictionary sv "support".) No incon­sistencies exist between the wording of the specification and these definitions and if it be borne in mind that support can

9

be vertical or lateral, more need not be said about the meaning of "self-supporting".

Another term in claim 1 that conveniently can be dealt with at this stage, is to be found in integer (j). That integer states that the self-supporting properties may be provided by additional welded stiffening zones "if desired". The appellant's case is that because of these words the claim does not sufficiently and clearly define the subject matter for which protection is claimed and is for that reason invalid. MacArthur J held that the effect of the option given by the words is to reduce integer (j) to an inessential integer. The result of his finding is that the monopoly covers not only a filter with, but also one without additional welded stiffening zones. Put differently, the claim can be divided nationally into two claims. Mr Franklin for the appellant attacked the finding and relied upon the reasoning of van Dijkhorst J in Ian Fraser-Johnston v G I Marketing CC 1993 BP 461 (C of P) at 477 F-G. He held that an integer in a dependent claim which is

10

"preferably provided" renders the claim unclear because one does not know whether the integer is part of the claim. (This judgment was confirmed on appeal, but the finding on this aspect of invalidity was not in issue: G Z Marketing (CC v Fraser-Johnston 1996 (1) SA 939 (A) 942F-G.)

There is little, if anything, to be gained by comparing the use of similar terms in different contexts to find the meaning or effect of one of them. Nor is there any justification for enquiring whether van Dijkhorst J's judgment was, in the circumstances of that case, correct or not. The question remains whether integer (j) renders this claim uncertain or unclear. In my judgment it does not and I fully agree with MacArthur J that the formulation used merely indicates that the integer is inessential in the sense described.

Infringement:

To prove infringement, the patentee relied upon the

11

evidence of Mr Reuvers who said that he found all the integers of claim 1 in the appellant's product. I have already mentioned that the appellant admitted that it had copied the patentee's commercial product. The appellant sought to establish that neither its product nor that of the patentee fell within the scope of claim 1 because, so it was said, the self-supporting properties of both filters were derived from the nature of the filtering material and not from the weld (integer (g)) or the spacer elements (integer (h)). To establish the proposition and to gainsay Reuvers' evidence, the appellant relied upon experiments conducted by Messrs Couchman and Mostert.

Reuvers' evidence is subject to the criticism that his opinion that the self-supporting property of the appellant's product was derived from the weld and the spacers was not supported by reasons. His testimony can be read to mean that because the pocket is self-supporting and has welds and spacers, the self-support derives from the welds and

12

spacers. If he intended to say so, he was mistaken. The specification does not promise, nor is it a scientific fact, that welds and spacers used on filter pockets of a suitable material inevitably produce a filter with self-supporting properties.

On the other hand, I have some difficulty in appreciating the relevance of the appellant's experiments. A filter element without spacers was made of the same material as the allegedly infringing filter, but instead of being welded the pockets were sewn some distance from their edges and a blanket stitch was sewn around the border formed. Weights were attached to both elements and the resistance of the experimental model to sagging was compared to that of the appellant's commercial product. The latter gave way under less weight than the former. From this one can deduce that a filter pocket with the particular stitch on the particular material has greater self-supporting properties than one with welds and spacers. That does not prove that the self-supporting

13

characteristic of this product is not the consequence of the welds and the spacers, whether assisted by the nature of the material or not. As counsel for the appellant repeatedly and correctly stressed, if the self-supporting ability derives from something in addition to the welds and spacers, infringement is not avoided. In view of Couchman's concession that the welds and spacers, to a limited extent, support the appellant's filter element laterally, having regard to the principle just stated and the meaning of 'self-supporting' dealt with earlier, it follows that the court below correctly found for the patentee on the question of infringement of claim 1.

Claim 5, according to Mr Moldow, an expert called by the appellant, was also infringed and I did not understand Mr Franklin for the appellant to seriously contend otherwise. This claim differs from claim 1 inasmuch as it does not require the pockets to have a self-supporting property provided by welds and spacers although the pockets are welded and have spacers. These, according to the specification, are for

14

stiffening and stabilising the pockets and to provide optimum inflow conditions. The welds, the specification says, provide a proper seal and nothing more. It is unnecessary to dwell any further on the subject because it was not shown that MacArthur J had erred in holding this claim to have been infringed.

Invalidity - general:

I have disposed of "ambiguity" of claiming while dealing with the interpretation of the specification. Inutility, lack of novelty and lack of subject-matter (obviousness) remain. It is trite that each objection must be considered in isolation, but that does not mean that evidence led on one aspect may not be relevant on another. Furthermore, a patent has one meaning only and it applies to both infringement and validity.

Inutility:

An invention is not useful if it does not

1 5

effectively produce the result aimed at or promised by the specification. (See B-M (Group (Pty) Ltd v Beecham Group Ltd 1980 (4) SA 536 (A) 550C-551D; Selero (Pty) Ltd and another v Chauvier and another 1982 (2) SA 208 (T) 212 upheld and approved on this aspect: Selero (Pty) Ltd and another v Chauvier and another 1984 (1) SA 128 (A) 143D-E.) The promise is usually to be found in the body of the specification and its meaning is a matter of construction (cf B-M Group at 551C).

I dealt with the promise of the invention and found that the specification promises that a filter made according to the claims attains the performance of the then known rigid high surface filters and avoids the disadvantages of the prior art filters, especially fluttering, bulging, difficult replacement, and breakage of the material.

The appellant did not allege that these promises were not fulfilled by the claimed invention. What the appellant sought to prove was that "a continuous edge trimmed weld and spacer elements secured to the filter medium ... and ...

16

additional welded stiffening zones do not provide the pockets with self-supporting properties." What was established was that not all filters with spacers and weld are self-supporting and that not all self-supporting filters have spacers and welds. Since the evidence and argument were based upon a misconception of the promise of the invention the attack on the utility of the patent fails.

Lack of novelty - introduction:

The appellant relies upon several separate allegations for its attack upon the novelty of claim 1 and the claims dependent upon it. (Claim 5's invalidity was not raised during the trial or the course of argument and the claim will not feature any further in this judgment.) These are:

1. that the invention was described in a publication -referred to as "BB" - of which there was a copy in the Republic at the effective date, 17 September 1975;

2. that the invention was used in the Republic before the

17

effective date or known to -

(i) Mr Bubenzer of Brandt Engineering (Pty) Ltd (referred to as "Brandt");

(ii) H Lewinberg (Pty) Ltd;

(iii) Mr Ubsdell of Air Vent Services, and

(iv) the Transvaal Department of Works; (c) that the invention was published in print during March-April 1975 in Germany in two brochures referred to as "A" and "B". (See the definition of "new" in s 1 (ix) of the 1952 Act.)

The patentee objected to the reliance on "A" and "B" and the objection caused the filing of a petition by the appellant shortly before the hearing of the appeal in which it applied for an amendment to its so-called "Response to Pre-trial Questions". The reason the appellant requires the relief appears from the facts that follow. In the counterclaim for revocation, the appellant relied upon many alleged anticipatory documents, including "A" and "B". Its expert notices were,

18

however, confined to a discussion of "BB". The particulars for trial, on the other hand, stated that the appellant would rely on "BB" and yet another document, "AA". Because of the confusion caused by these inconsistencies, the appellant was requested at the pre-trial conference for clarification. The appellant's reply was that the attack was limited to "BB" as the only anticipatory document. In a written response to pre-trial questions, the appellant reaffirmed its position. During the opening address, counsel for the patentee informed the court of the limitation of the issue to "BB". This statement did not elicit any response from the appellant's counsel. In spite of this, he argued at the conclusion of the trial that "A" and "B", besides "BB", anticipated the claimed invention. He now wishes to do the same in this Court.

Mr Mostert, on behalf of the appellant, alleged in his affidavit that the limitation of the issue was the result of confusion caused by the nature of the question posed. This excuse cannot, in the light of the facts recited, be accepted.

19

To allow a party, without special circumstances, to resile from an agreement deliberately reached at a pre-trial conference would be to negate the object of rule 37, which is to limit issues and to curtail the scope of the litigation (cf Price NO v Allied-JBS Building Society 1980 (3) SA874 (A) 882D-H). If a party elects to limit the ambit of his case, the election is usually binding (AJ" Shepherd (Edms) Bpk v Santam Verseteringsmaatstappy Bpk 1985 (1) SA399 (A) 415B-D; Chemfos Ltd v Plaasfosfaat (Pty) Ltd 1985 (3) SA 106 (A) 114I-115B). No reason exists why the principle should not apply in this case. The relief sought in the petition must consequently be refused.

Lack of novelty - document "BE":

To succeed on this ground the appellant has to establish (see s 1 (ix) (d) of the 1952 Act) that at the effective date -

(a) the document "BB" was published,

20

2. there was a copy of it in the Republic,

3. and the document describes the invention.

To prove (a) and (b) , the appellant relied on a compliments slip dated 15 September 1975, addressed to Mr Ubsdell, to which "BB" (dated August 1975) was attached. The compliment slip was that of Brandt and not of the patentee. Ubsdell no doubt received the compliment slip with "BB", but it is not possible to date the receipt. "BB", according to other Brandt documents was replaced by another brochure dated March 1976. During preparation for the trial other copies of "BB" were found in the possession of former clients of Brandt. Since the documentation relied upon purportedly emanated from Brandt, I have some difficulty in comprehending how the facts (especially the dates) they record are admissible against the patentee. Providing a copy to one person hardly amounts to a "publication", although it shows a willingness to publish. I should pause to point out that the patentee sought to discredit "BB" and the compliments slip, contending that they were not

21

genuine. Reason to suspect these documents initially there certainly was, but the patentee's attack on them failed completely. It is not necessary to reach a firm conclusion on whether the facts recited prove point (a) and (b) because of my conclusion on (c). On the other hand, the facts recited do have an important bearing on prior knowledge and use, subjects I deal with later.

Document "BB" is a technical pamphlet dealing on the face of it with a particular Viledon filter, stated to be a product of Freudenberg. The filter is identified as "Pocket Filter Type G 35/K", but the document does not state what G 35/K signifies. Technical data accompanied by a graph with no descriptive importance or of any relevance in the present context are provided. There is a photographic representation of the product and the text identifies its uses and sets out the following under four bullet points:

* "Homogeneous connection of frame to filter pockets by hard polyurethane foaming process."

22

* "Stable filter pockets are welded to form leak free
seams."

* "Total media usage due to reliable spacer elements."
* "Universal seal arrangement."

Netlon Ltd and another v Pacnet (Pty) Ltd 1977 (3) SA 840 (A) at 861H-862B, summarized the principles involved:

"... the defence (or objection) of anticipation relates to the claims and not to the description of the invention in the body of the specification in suit (see, too, the Letraset case, supra at pp. 264-5); the prior printed publication alleged to be anticipatory must be construed, for the exercise is primarily one of construing and comparing the two documents; moreover it must be construed as at the date of its publication to the exclusion of information subsequently discovered; the question then considered is whether the prior publication 'describes' the invention in suit as claimed; that is, whether it sets forth or recites at least the latter's essential integers in such a way that the same or substantially the same process or apparatus is identifiable or perceptible and hence made known or the same or substantially the same product can be made from that description in the prior publication; if the description in the prior document differs, even in a small respect, provided it is a real difference, such as the non-recital of a single essential integer, the anticipation fails; the opinions of expert-witnesses that the prior

23

publication does or does not anticipate a claim in suit must be disregarded for that is for the Court to decide."

A test referred to in Gentiruco AG v Firestone SA (Pty) Ltd 1972 (1) SA 589 (A) at 646F-G was applied (at 866E) -

"Hence for it to 'describe' the invented process etc., it must set forth or recite at least its essential integers in such a way that the same or substantially the same process is identifiable or perceptible and hence made known, or the same or substantially the same thing can be made, from that description. 'Substantially the same' means practically the same, or, to use Lord WESTBURY'S phrase adopted by VESSELS, J.A., in Veasey's case, p. 269, the same 'for the purposes of practical utility'; i.e., substance and not form must be regarded."

Letraset Ltd v Helios Ltd 1972 (3) SA 245 (A) 267A-C, emphasized that the court must look at a document through its own eyes and not through the eyes of a skilled worker in the art. In support of the proposition, Jansen JA quoted what he called an instructive passage from C van der Lely NV v Bamfords Ltd 1963 RPC 61 (HL) at 71 lines 5-19 (per Lord Reid) :

24

"We were informed that this is the first case in which the question of anticipation has turned on the proper inference to be drawn from photographs, although there have been cases of anticipation by published drawings. There is no doubt that, where the matter alleged to amount to anticipation consists of a written description, the interpretation of that description is, like the interpretation of any document, a question for the court assisted where necessary by evidence regarding the meaning of technical language. It was argued that the same applied to a photograph. I do not think so. Lawyers are expected to be experts in the use of the English language, but we are not experts in the reading or interpretation of photographs. The question is what the eye of the man with appropriate engineering skill and experience would see in the photograph, and that appears to me to be a matter for evidence. Where the evidence is contradictory the judge must decide. But the judge ought not, in my opinion, to attempt to read or construe the photograph himself; he looks at the photograph in determining which of the explanations given by the witnesses appears to be most worthy of acceptance."

Mr Franklin used the quotation for another purpose, namely to impress upon the Court that it should refrain from interpreting the photograph in "BB" but should be led by what the witnesses had to say about it. I will limit myself to say that my impression is that lawyers are no longer expected to

25

be experts in the use of any language; photographs - moving or still - form part of everyone's daily diet but seldom require much by way of expertise to enable them to be "read"; if anything, I would underline the concluding words.

Mr Moldow, the appellant's expert, when shown the photograph on "BB" and having his attention drawn to a black line inside the filter pockets was asked the leading question -one of many - whether he saw a weld line. Moldow did not accept the suggestion and identified the line as a small thin steel rod or wire cast into the support frame. He expressed the opinion that the filter pockets shown were not self-supporting.

Undaunted by this evidence, Mr Franklin argued that the rod (which is patently present) does not support the filter depicted on the photograph. His submission was that the text states that the filter pockets are welded and the photograph shows the presence of spacers; the patent implies that filter pockets that are welded and have spacers are self-supporting;

26

the filter shown with the rod removed would therefore also be self-supporting; the rod is simply an additional integer and should be ignored when considering prior description.

The fallacy of the argument was exposed in the context of infringement and inutility where it was pointed out that it is premised upon an incorrect construction of the patent. To repeat, there is no evidence that a filter with welds and spacers inevitably has self-supporting properties. Nothing in "BB" suggests that the filter will not sag or collapse under its own weight if the rod is removed. This means that "BB" fails to pass muster.

Anticipation - prior use or knowledge:

An invention was also not 'new' under the 1952 Act if the invention was "known or used in the Republic by anyone other than the applicant or his agent ... (secret knowledge ... being excluded)". (See s 1(ix)(a)). Although the appellant pleaded the individual uses or knowledge referred to earlier,

27

the evidence should rather be considered as a whole to judge its effect. The major problem confronting the parties was that the facts to be established or disproved occurred some twenty years ago. Witnesses were bound to rely on imperfect memories, often prompted by what they had heard from others, or by incomplete documentation. Such evidence is inherently suspect. Honesty is not the issue, rather the reliability of the evidence. Moldow, for instance, felt the need to apologise for his failing memory and conceded that he had no independent recollection of the events which took place during 1975. He did not know of the patent before the 1980 's and he had no reason to remember the sequence of events. Bubenzer was to recall what had happened at a company he had left in 1979. The company, Brandt, was liquidated during the intervening years and some of its documents - clearly not all - were located a few days before the trial by the appellant. The patentee did not have free access to these documents. The patentee, it was accepted, did not retain any records relating to the relevant

28

period. In such circumstances the onus is usually decisive and one tends to seek for corroboration rather than to rely on the ipse dixit of honest witnesses.

The patentee was the manufacturer of a filter material marketed under the trademark "Viledon". Because of serious competition, it decided to develop a pocket filter using the Viledon material. Development took time. On 18 March 1975 filter design technicians within the VILEDON group held a conference attended by Moldow. Moldow, then the Danish agent of the patentee, was a member of the development team and visited the patentee regularly about this over a period of more than two years. At the meeting production samples of "the now completed COMPACT 'GROB' and 'FEIN' Pocket Filter Elements" were displayed and the meeting was informed that they were to be called "Viledon Compact Filter Bags". ('Grob' is German for 'coarse' and 'fein' means 'fine'; both terms describe the nature or grade of the Viledon filter material.) The minutes of the meeting do not give the physical features of the

29

products in any detail, but state that they are "very compact, internally stable, 'flutter-proof ... [have] the homogeneous 'filter-frame' connection, [and] the functionally reliable spacer element etc ...". Attached to the minute is a page setting out the technical data of the two filter bags. 'Grob' has four pockets and its technical data are identical to those of the product shown in "BB".

Shortly after the meeting, and at a trade fair in Germany, the patentee exhibited its filters behind closed doors to a select group of customers. A. circular (probably addressed to the patentee's agents) of 7 April 1975 offered the Viledon compact to the recipients. Two weeks later a short pamphlet (document "B" referred to earlier) was sent to Moldow and he was asked to state the number of copies he required. Document "B" is fairly indistinct but it does contain a representation of two "Viledon compact filter bag unit[s]", the one 'grob' with four pockets and the other 'fein' with eight. Much more is not visible and although it states that the filters are

30

stable and self-supporting, it gives no reason for or explanation of this quality. The technical data does not differ from those given in the conference minutes, and, for "grob", in "BB".

These products were soon marketed because it is known that by 3 September 1975 they had 'found full acceptance in the marketplace' and 'interesting turnovers' had been attained. They reached the South African market in unknown quantities. That was established at the trial through, at least, Mr Ubsdell who during July 1975 purchased from Brandt, the patentee's local agent, 'Viledon G35 Coarse' and 'Viledon F45 Fine' filters. As mentioned, document "BB" dated August 1975 was sent to Ubsdell without any restrictions two days before the effective date of the patent.

On 3 September 1975, the patentee informed its agents that it had a new pocket for the market, namely a Viledon Compact "Grob 8", a "new version with 8 pockets". The existing pockets were to be marketed as Viledon Compact "Grob 4" and

31

"Fein". The first reference to these three names in trade literature is found in a technical brochure of Brandt dated March 1976 which, according to a note found, replaced "BB". The filters shown in this brochure do not, on the face of it, have any rods. The technical data concerning 'Grob 4' differ in some respects from that contained in the product it was supposed to replace, namely the one in "BB". Technical differences between the "Fein" disclosed at the meeting of 18 March 1975 and the "Fine" of the March 1976 document are apparent. There is no indication at all when the filters offered in the circular of 3 September 1975 reached the market, whether in South Africa or elsewhere. What is relevant, is that all the sales relied upon by the appellant as proof of prior knowledge or use in South Africa predate this circular. On what is on record, these filters had to be those described in the minutes of 18 March 1975, one of which was depicted on "BB". Nothing further is known of the "Fein".

To prove that the Viledon pocket filters brought to

32

South Africa before the effective date were made according to the invention, the appellant relied heavily on what was submitted to be common cause, namely that (quoting from the heads of argument) the 'compact filters of 1975 were identical to those of today and that they had all the integers of the claims' and that 'it was admitted by the respondents that the first respondent has only ever sold one type of Viledon compact filter, i e that made in accordance with the patent'.

Temporary interdict proceedings preceded the trial, and in a replying affidavit a deponent - the head of the patentee's patent department - who did not testify at the trial, although present, admitted that 'the configuration as well as the production method of today's VILEDON COMPACT FILTERS is essentially identical to those of 1975. There were only some minor changes in the quality of the materials and fibres utilized.' He added that the Viledon bag filters were self-supporting and that filters were produced fitted with a cross-rod to provide additional support for the filter pockets.

33

The date given, namely 1975 is, in the light of what 1 have said, somewhat ambiguous and confusing. Reliance was also placed upon the evidence of Mr Schorn, who, at the time was employed by Brandt. According to his testimony, the patentee, to his knowledge, never made any Viledon compact filters save according to the invention. There is nothing to show that he knew the contents of the specification, either in 1975 or even at the time of the trial. Finally, Bubenzer, the manager of Brandt at the time, asked whether the patentee had ever sold any pocket filter other than the pocket filters exemplified by the exhibits before the court, said no.

Many exhibits were before the court; some of them did not fall within the scope of the patent. For instance, there was another version of the Viledon G35/K (which first appeared on "BB") without a rod, spacers or welds for the spacers and with five pockets. There were Viledon M 85 and M 95 pocket filters that collapse if their holding wires are removed. These Viledon pocket filters that fall outside the scope of the

34

patent were introduced by Mr Franklin to prove the lack of utility. Another Viledon filter not covered by the patent is the one exemplified on "BB". Mr Franklin thus clearly and effectively established that the 'admissions' he wishes to rely upon were wrongly made. His submission that the matter was common cause, overstates the position, because it was the appellant's main case that neither the ordinary Viledon filter, nor the appellant's copy, falls within the scope of the claims because the self-supporting properties derive from the filter material and not the welds and spacers.

Moldow did not take the matter much further. He knew the product depicted in "BB" and although he said that it was an early model (which it was), he did not say (as Mr Franklin put it to Schorn and during argument) that the product was merely a prototype. Ubsdell's evidence proves no more than that he purchased Viledon pocket filters and requested and received "BB". He did not say that the filters received were identical to or different from that portrayed on "BB". In an

35

attempt to remedy this defect in his evidence, Mr Franklin sought to rely on Ubsdell's affidavit filed by the appellant in the interim interdict proceedings in which he expressed the opinion that the filters he had dealt with during 1975 had all the integers of claim 1. He did not describe or illustrate the filters, he gave no reasons for his opinion, he was not called as an expert, and he did not confirm his opinion in court. The belated recourse to s 34(1) of the Civil Proceedings Evidence Act 25 of 1965 must fail. No intimation was given at the trial that reliance would be placed upon the contents of the affidavit in addition to the oral evidence. Nor was any reason proffered why Ubsdell could not have given the evidence in court. In short, 'there was no factual reason or need for the appellant to invoke sec. 34 at all' (Narlis v South African Bank of Athens 1976 (2) SA 573 (A) 578A-B).

Only the evidence concerning the prior use at or by H Lewinberg (Pty) Ltd deals in any way with the question whether the filters that were sold before the effective date

36

fell within the scope of the claims. MacArthur J found the evidence on the issue 'unconvincing and somewhat contrived' and I am not satisfied that his assessment is wrong. On the assumption that he was too critical, 1 do not think that one can rely, without a smattering of corroboration, on the statement of Mr Mostert that the filter he saw during January 1976 had no rods or wires. Even if one could, the evidence does not show, directly or by way of inference, that this filter was supplied and installed before the effective date. The witness who saw Viledon filters on the site before the effective date, Mr Schofield, was very vague, understandably so, on what he had seen, and there is no reason why Mostert should have had a clearer recollection of what, at the time, was an insignificant aspect of an otherwise mundane event.

It needs to be recorded that before the issue was raised during argument in this Court, counsel for the patentee, perhaps blinded by the dust of the arena, failed to comprehend the import of the facts as they unfolded during the trial. It

37

is a valid criticism of the patentee's case that it never sought to establish the nature of the Viledon filters as it was before the effective date of the patent. The case the patentee set out to make was that Viledon filters could not have been on the market in South Africa before the effective date. The attempt was unsuccessful. However, the stance taken by the patentee at the trial is understandable if regard is had to certain discrepancies in the documents and evidence, the late discovery of documents found under strange circumstances by the appellant, the liquidation of the patentee's South African agent (Brandt), the lapse of time, the absence of full and proper documentation and the unreliable memory of witnesses whose honesty was not impeached. Nevertheless, the criticism raised is an important factor which should be placed in the scale in deciding the probabilities.

Once the finding is, as it has to be in the light of all the evidence, that not all Viledon bag or pocket filters necessarily fell within the scope of the patent, one cannot

38

conclude with any degree of conviction that the filters known or sold in South Africa (or for that matter, abroad) before the effective date fell within the scope of the claims of the patent. That disposes of the attack upon the novelty of the patent.

Obviousness:

The alleged lack of subject-matter was based upon the evidence of Moldow, but his evidence on the issue was destructive of the appellant's case. According to him, there was an urgent commercial need for the product because he was unable to compete in the marketplace. The patentee began with research work during 1973 and allocated a group of technicians to the project. Moldow, an expert in the field, acted as consultant. The project was successfully completed only during 1975. The difficulty the patentee had in finding a solution for the known problem, given his evidence that all the integers of the invention were known to him as an expert, belies any

39

suggestion that the invention could have been obvious. Nor does his evidence answer the question whether the invention (the specific combination and interaction of the integers) was obvious to the ordinary worker in the art.

I find it unnecessary to deal in any detail with the evidence on the matter because it was not shown that the trial judge had erred in his assessment. The only point that needs mentioning, is the argument that the viledon filter was part of what was common knowledge at the effective date of the patent. Assuming in favour of the appellant that the filters referred to in the minutes of 18 March 1975 indeed became part of the common knowledge before the effective date, the question remains what those filters were, and if the appellant could not discharge its onus on this question under the heading of prior knowledge and use, it similarly failed to do so under the present rubric.

Costs of the interim interdict proceedings:

40

The patentee applied for a temporary interdict which was not disposed of and the costs were reserved, either by

agreement or by court order, for the trial court. They were argued and MacArthur J, without supplying any reasons, ordered the appellant to pay those costs. One can assume that he based his decision on the reasoning that the appellant, having lost the main case, had no ground for opposing the interdict application. This was the approach Mr Ginsburg submitted this Court should adopt. Mr Franklin, on the other hand, submitted that the patentee should have been ordered to pay the costs because the application was not proceeded with, should never have been launched and that a false reason was given why the application was not moved. These arguments, and others, were addressed to MacArthur J and there is nothing to show that he did not have due regard to them. He was in the favourable position that he had the full application record whereas only portions of the record were placed before this Court by the appellant. Even the notice of motion is missing and, for

41

example, only pages 1, 10, 14 and 15 of the founding affidavit are in the record. This Court is therefore not in a position to reconsider the exercise of his discretion.

In the result —

1. the application for amendment of the "Response
to Pre-trial Questions" is dismissed with costs;

2. the appeal is dismissed with costs;

3. the costs in par 1 and 2 include the costs
consequent upon the employment of two counsel.

L T C HARMS JUDGE OF APPEAL

EKSTEEN JA )

FH GROSSKOPF JA) CONCUR

SCHUTZ JA )

 

 

 

 

 

 

CASE NO: 258/96

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

In the matter between:

FILTA-MATIX (PTY) LIMITED Appellant

and

CARL FREUDENBERG First Respondent

FREUDENBERG NONWOVENS Second Respondent
(PTY) LIMITED

FILTRA (PTY) LIMITED Third Respondent

CORAM: Eksteen, FH Grosskopf, Harms, Marais ef Schutz JJA HEARD: 13 & 14 November 1997 DELIVERED: 27 November 1997

JUDGMENT

MARAIS JA/

2

MARAIS JA:

I have had the benefit of reading the judgment of my brother Harms. I have had considerable difficulty in deciding the anticipation point. When the circumstantial evidence which the appellant was able to assemble in difficult circumstances after so long a lapse of time is coupled with the direct evidence of prior knowledge or use, it cannot be easily dismissed as unimpressive. The respondents' response to that evidence was not that the filters which were sold prior to the critical date were not the relevant filters but some other filters. Its dogmatic response was that no filters had been sold in South Africa other than those to which the patent relates, that they were sold after the critical date, and that any documentation suggesting the contrary was a forgery. That remained its stance throughout a long and protracted trial. That stance was shown to be untenable and in my view, there is a good deal to be said for the

3

proposition that the appellant's case was strengthened by the failure of the only factual defence put forward by the respondents. Whether the appellant has discharged the onus of proof which rests upon it is the ultimate question. The factors which have been listed in the judgment of Harms JA as justifying the conclusion that the onus of proof has not been discharged are certainly relevant factors but I remain troubled by the respondents' steadfast refusal to put up such a case and its insistence, even before this court, that the only filters sold in South Africa were those to which the patent relates, and that none to which it did not were ever sold here. However, I do not feel so convinced that my view of the matter is right that I feel obliged to dissent from the judgment of the majority. Dubitante therefore I concur in the dismissal of the appeal. I should add that in all other respects I concur unreservedly in the judgment of the majority.

R M MARAIS