OUTER HOUSE, COURT OF SESSION
 CSOH 114
OPINION OF LORD GLENNIE
in the cause
SCOTNET (1974) LIMITED and OTHERS
Pursuers: Currie, Q.C., Robertson; McClure Naismith
Defenders: Moynihan, Q.C., Frame-Bell; Burness, W.S.
 In this intellectual property cause I heard a preliminary proof before answer restricted to issues of infringement. As matters developed during the course of the proof, certain issues, which had been live during the pursuers' evidence, fell away. In the event, therefore, the ambit of the dispute narrowed considerably, essentially to a question of construction of the pursuers' patent.
 For the purpose of their closing submissions, both parties lodged written submissions, for which I am grateful. These were taken as read; and in consequence parties were able, in their oral argument, to focus on the issues that the written submissions had thrown up.
 The pursuers and defenders both manufacture elasticated tubular netting of the type used for wrapping joints of meat. They do so using small diameter circular warp knitting machines. The pursuers are the registered owners of GB Patent 2333301B ("the 301 Patent"), which patented certain improvements in such machines. They use machines manufactured in accordance with that patent. I shall refer to the pursuers' machine as the "Trunature machine". The defenders use a machine known as the "Turbo 98". The pursuers allege that the defenders' machine infringes certain claims in the 301 Patent.
Operation of a circular warp knitting machine
 I take the description of how a circular warp knitting machine works from the Report by Mr Adrian Chettle, a mechanical engineer by training and now a Chartered Patent Agent and European Patent Attorney, who gave evidence for the pursuers. I did not understand this description to be controversial.
"In a circular warp knitting machine the warp threads run generally vertically down and are arranged in a circle around a cylinder. Several pressers are driven around the machine at high speed and knit successive warp threads on reciprocating needles, one needle being provided for each warp thread. The separate knitted warp chains are drawn down out of the centre of the cylinder under tension, and are endless so long as the thread supply is maintained.
The rotating pressers periodically urge successive warp threads radially inwardly to be engaged by the respective needles. A typical machine has four pressers to form chains of successive warp thread loops.
A weft thread feeder may rotate with the pressers so that plain loops are knitted between periodic weft engaging loops. The number of plain loops is one less than the number of pressers for a single weft thread. In this form the knitting machine will knit a continuous sock by inlaying or knitting the weft thread to the warp threads. Several weft thread feeders may be provided. The weft may be laid from the inside or from the outside."
The pressers typically rotate anti-clockwise around the cylinder.
 The key to the successful operation of the knitting machine is to ensure that the warp threads are consistently engaged by the reciprocating needles. Typically, the needles are moved up and down my means of cams, which rotate around the cylinder with the pressers. The needles have an open hook at the upper end and a hinged latch by means of which the hook is opened and closed. The latch falls open as the needle rises; the warp thread is deflected into the open hook by a rotating presser; and as the needle descends, having caught the thread, the latch is closed by the movement of the previous loop of thread. This process is repeated endlessly so long as there is enough thread. It is important that the warp thread is deflected into the open hook for a period long enough to ensure that it cannot escape before being trapped in the hook by the descending needle. However, the thread need not necessarily be pressed inwardly into the needle hook for the full period before the latch is closed because, at some point before the latch closes, the needle will have descended sufficiently to enable the hook to have trapped the thread securely.
The 198 Patent
 The pursuers' patent sought to improve upon machines then in use, a typical one being that described in UK Patent 2214198 ("the 198 Patent"). According to the specification of the 198 Patent, the descending warp threads pass through wire guides fixed to the cylinder in a position just above the highest point reached by the hook of the needle. These wire guides were referred to during the proof as "staples". Their purpose was to hold the thread in position and, in conjunction with the rotating presser plate, guide it into position in way of the open hook of the needle. Their shape can be described in this way. The leading edge of the staple (i.e. the edge nearest the presser plate as it approaches in an anti-clockwise rotation) protrudes from the curved surface of the cylinder at the normal (i.e. at an angle of 90º to the tangent). The wire then turns at an acute angle before returning in a curve to the cylinder. The warp thread, when not being pressed, lies against the straight edge of the staple. When the rotating presser plate moves into contact with the thread, it pushes the thread onto the curved part of the staple, and that pressure on the thread against the curvature of the staple guides the thread inwards towards the cylinder. By this means the thread is pressed into a position where it brushes across the face of the needle below the hook and is caught by the hook of the needle as it descends. As the presser moves further around the cylinder, it releases the thread. The released thread remains held within the staple.
The 301 Patent
 The pursuers' managing director, Stuart Revill, who gave evidence at the preliminary proof, was the inventor named in the 301 Patent. According to the descriptive part of that patent, the provision of the staples in the 198 Patent limits the number of needle tricks which can be provided in a given cylinder. The needle trick is the groove within which the needle moves up and down, and limiting the number of needle tricks limits the number of needles involved in the knitting operation. The descriptive part of the patent goes on to state that the applicants
"have discovered that if warp threads approach the cylinder at an appropriate angle and if the presser plate is appropriately shaped and positioned there is no necessity to surround each warp thread with such a guide [viz. a staple]. Dispensing with these guides permits the use of up to twice as many needles in the same machine, thus permitting the manufacture of a more closely or densely knit product without any detriment to the speed and efficiency of the operation. The machine of the present invention is simpler, utilises less components and is easier to maintain".
There is then mention of another improvement, namely the latch guard, which prevents the latch being cast off too abruptly and closing at the wrong time. In the Trunature machine the latch guard takes the form of a wire running horizontally between and below the presser plates and rotating with them.
 Against that background, the principal claim in the 301 Patent (Claim 1) is as follows:
"1. A high speed circular warp thread knitting machine of the kind in which latch needles are reciprocated in respective, vertical tricks in a stationary cylinder by rotating cam means thereby to engage warp threads drawn down from above, a presser plate rotating with a said cam means being provided to deflect each warp thread in turn inwardly of the cylinder across a needle while the latch of the latter is open, wherein the warp threads are drawn through respective guides in a stationary element mounted above the cylinder, the angular relationship of said element to the cylinder being such that each guide is angularly offset relative to the associated needle trick in the direction opposite to the intended direction of rotation of the presser plate, the presser plate having a leading surface presented towards the cylinder at an acute angle to a radius of the cylinder and the arrangement being such that, when not deflected by the presser plate, each warp thread passes inwardly of the cylinder in a straight line at an acute angle to the vertical between the associated guide and the top of the associated needle trick without passing through or around any other thread guidance or deflection means."
I should also quote Claim 5, which is in the following terms:
"5. A machine as claimed in any one of the preceding claims, wherein a latch guard is provided positioned to obstruct closing movement of the latch of an ascending needle just after the latch has cast off onto the stem of the needle a previously formed loop."
The Turbo 98 machine
 The defenders' Turbo 98 knitting machine also operates without any wire guides or staples attached to the cylinder to guide the thread. The difference, which for present purposes is the only difference which I need mention, between the Trunature machine and Turbo 98 machine lies in the part which performs the function of deflecting the yarn around the needle. In the 301 Patent, as we have seen, this is called the "presser plate", whereas in the Turbo 98 machine it is known as the "placer".
presser plate and the placer
 The presser plate on the Trunature machine is a small steel plate. It's shape is approximately rhomboid, with sides of approximately 21/2 cm in length, and angles of approximately 100º and 80º; but with this difference, namely, that at one of the oblique angles the corner has been taken off, so that there are, in effect, five surfaces. The operative surfaces are the "leading edge", which is at an acute angle to the cylinder and contacts the yarn to begin deflecting it; the "chamfered surface", which is the surface cutting across what would otherwise be the fourth corner of the rhomboid and is positioned so as to run nearly parallel to the tangent at the nearest part of the cylinder; and the "trailing edge" which holds the yarn in the hook for long enough for it to be secure before releasing it.
 The Turbo 98 placer is significantly larger (over 10 cm long) and differently shaped. There was lodged in process a drawing of the placer divided into three sections. Parties were agreed that, depending upon precisely where one drew the lines notionally separating it from the two end sections, the central section ("D") of the placer was the equivalent of the Trunature presser plate. The foremost part of the placer ("B") was known as the "pre-tensioner"; and the aftermost part ("C") as the "post-tensioner". Sections B, D and C lie in the horizontal plain. The latch guard on the Turbo 98 machine, more substantial than on the Trunature machine, is fixed directly below the three sections of the placer.
 The defenders contended during the proof that the pre- and post- tensioners served a useful function. According to them, the function of the pre-tensioner was to produce some slack in the yarn before it was taken up by the central section D; and to place the yarn in the correct position against the leading edge of the central section. The function of the post-tensioner was to release the warp thread smoothly after it had been caught in the hook of the needle, thereby avoiding vibration or "twanging" which might cause the next stitch to be missed; and to ensure that the warp thread passed smoothly over the weft yarn feeder. In light of events which happened at the proof, I only heard partial evidence on some of these points. On that evidence, I am satisfied that although the pre- and post-tensioners deflect the yarn from the straight, neither performs any useful function relative to the knitting process. There is no need for the tension in the yarn to be slackened before being presented to the leading edge of the central section of the placer. Nor is there any need to position the yarn to the pre-tensioner; if properly set, the machine delivers the yarn to the correct position without this. It was not shown that the gradual release of tension by means of the post-tensioner achieves anything useful, in that there was no evidence that the Trunature machine, which does not use a post-tensioner, has a propensity to drop stitches. So far as the warp feeder is concerned, it seemed clear to me that by holding the yarn deflected towards the cylinder for longer than was necessary, the post-tensioner holds the warp yarn in the path of the weft yarn feeder; so that, when the weft yarn feeder moves around the cylinder, the action of the post-tensioner creates increased friction between the warp yarn and the weft yarn feeder. It certainly did not help the warp yarn pass smoothly over the weft yarn feeder. For these reasons, I do not accept the defenders' submission that the pre- and post-tensioner serve any useful function in the knitting process. I should also say at this point that I do not accept the defenders' case that the warp thread was deflected from the straight by the latch guard on the Turbo 98 machine. According to the evidence I heard, and the photographs and DVDs shown in evidence which recorded the machine in action, there was no contact between the latch guard and the thread. Any such contact, had it occurred, would have been undesirable.
 The pursuers claim that the defenders' Turbo 98 machine infringes Claims 1 and 5 of the 301 Patent. It is agreed, however, that the alleged infringement of Claim 5 is dependent upon the arguments in relation to infringement of Claim 1. Accordingly, the focus of the discussion before me turned on the proper construction to be given to Claim 1, and in particular the concluding part of Claim 1 which, for convenience, I set out again at this point:
"...the arrangement being such that, when not deflected by the presser plate, each warp thread passes inwardly of the cylinder in a straight line at an acute angle to the vertical between the associated guide and top of the associated needle trick without passing through or around any other thread guidance or deflection means." [emphasis added]
It is the defenders' contention that only the central part, D, of their placer is a "presser plate", as that term is to be understood in the 301 Patent. Accordingly, on their construction, the Turbo 98 machine does not infringe the monopoly claimed in the 301 Patent because the thread, "when not deflected by the presser plate", does pass around thread guidance or deflection means, namely sections B and C of their placer. (They also assert that the latch guard is another deflection means but, since I have found that there is no contact between the latch guard and the thread, I need not consider this aspect of the submission further.)
 The principles to be applied in construing the 301 Patent were not in dispute. Section 125 of the Patents Act 1977 provides that, for the purposes of the Act, an invention for a patent for which an application has been made or for which a patent has been granted "shall, unless the context otherwise requires, be taken to be that specified in a claim of the specification of the application or patent, as the case may be, as interpreted by the description and any drawings contained in that specification" and it goes on to provide that the extent of any protection conferred by the patent shall be determined accordingly. That section also applies the Protocol on the Interpretation of Article 69 of the European Patent Convention. In Kirin-Amgen Inc. v Hoechst Marion Roussel Ltd  R.P.C. 9, the House of Lords approved, as consistent with the 1977 Act and the Protocol, the classic approach to construction set out by Lord Diplock in Catnic Components Limited v Hill & Smith Limited  R.P.C. 183, 243:
"A patent specification should be given a purposive construction rather than a purely literal one derived from applying to it the kind of meticulous verbal analysis in which lawyers are too often tempted by their training to indulge. The question in each case is: whether persons with practical knowledge and experience of the kind of work in which the invention was intended to be used, would understand that strict compliance with a particular descriptive word or phrase appearing in a claim was intended by the patentee to be an essential requirement of the invention so that any variant would fall outside the monopoly claimed, even though it could have no material effect upon the way the invention worked."
 In Improver Corp v Remington Consumer Products Limited  F.S.R. 181, Hoffman J. referred to the approach of Lord Diplock in Catnic Components Limited v Hill and Smith Limited and, at page 189, set out what have become known as the "Improver" or "Protocol" questions:
"If the issue was whether a feature embodied in an alleged infringement which fell outside the primary, literal or a contextual meaning of a descriptive word or phrase in the claim ('a variant') was nevertheless within its language as properly interpreted, the court should ask itself the following three questions:
(1) Does the variant have a material effect upon the way the invention works? If yes, the variant is outside the claim. If no -
(2) Would this (i.e. that the variant had no material effect) have been obvious at the date of publication of the patent to a reader skilled in the art. If no, the variant is outside the claim. If yes -
(3) Would the reader skilled in the art nevertheless have understood from the language of the claim that the patentee intended that strict compliance with the primary meaning was an essential requirement of the invention. If yes, the variant is outside the claim.
On the other hand, a negative answer to the last question would lead to the conclusion that the patentee was intending the word or phrase to have not a literal but a figurative meaning (the figure being a form of synecdoche or metonymy) denoting a class of things which included the variant and the literal meaning, the latter being perhaps the most perfect, best-known or striking example of the class."
I should also mention a further passage from the judgment of Hoffman J in Improver at page 190:
"It is worth noticing that Lord Diplock's first two questions, although they cannot sensibly be answered without reference to the patent, do not primarily involve questions of construction: whether the variant would make a material difference to the way the invention worked and whether this would have been obvious to the skilled reader are questions of fact. The answers are used to provide the factual background against which the specification must be construed. It is the third question which raises the question of construction and Lord Diplock's formulation makes it clear that on this question the answers to the first two questions are not conclusive. Even a purposive construction of the language of the patent may lead to the conclusion that although the variant made no material difference and this would have been obvious at the time, the patentee for some reason was confining his claim to the primary meaning and excluding the variant. If this were not the case, there would be no point in asking the third question at all."
In this connection, Mr Moynihan referred me to a passage in the speech of Lord Diplock in Catnic at page 244, in which he had stressed the absence, in that case, of a "plausible reason ... why any rational patentee should want to place so narrow a limitation on his invention".
 I was reminded by Mr Moynihan that in Kirin-Angen, Lord Hoffman emphasised that the Improver questions were merely guidelines. Further, he submitted that they were an unnecessary distraction in the present case because the true question was to determine the proper interpretation of the patent; and, as the Improver case shows, the first two questions can be answered adversely to the defenders without there necessarily being any infringement. I accept that the true question is to determine the proper interpretation of the contract, rather than move too rapidly to (or focus too rigidly on) the Improver questions; and for this reason, that before one gets to the Improver questions, there is a prior question to be resolved, namely whether the Turbo 98 machine is a variant at all, i.e. (to adopt the formulation used by Hoffman J in Improver) whether it falls "outside the primary, literal or a contextual meaning" of the claim. There are two discrete arguments relevant to this question.
 The first raises the question whether the whole of the defenders' Turbo 98 "placer" corresponds to the Trunature "presser plate" or whether, on the other hand, the pre- and post-tensioners, sections B and C, are separate and distinct parts of the machine. The question is somewhat arcane, and may give of a variety of answers depending on the context in which it is asked. Here it is asked not in the abstract but for the purpose of understanding the defenders' machine in terms of an alleged infringement of the 301 Patent. To answer the question it is necessary to ask what, in terms of the 301 Patent, is meant by the "presser plate". Clearly it is not limited to a plate of the precise dimensions and angles of the Trunature presser plate. This was, as I understood it, common ground, otherwise the defenders' acceptance that the central part of their "placer" corresponded to the presser plate would make no sense. Mr Moynihan argued that one should look to the definition of "presser plate" in Claim 1 of the 301 Patent, which was to be found in the passage stating that the presser plate has "a leading surface presented towards the cylinder at an acute angle to the radius of the cylinder". He submitted that the Turbo 98 placer as a whole did not meet this description. The relevant surface of the placer, which corresponded to the leading surface of the presser plate, was the surface at the forward part of the central section D of the placer. Therefore section D was the equivalent of the presser plate, while the pre- and post-tensioners (sections B and C) were wholly distinct and were not part of the one presser plate.
 I do not see why the fact that the relevant "leading surface" is the front edge of the central section D should prevent the whole of the placer being a presser plate. According to the definition relied upon by Mr Moynihan, the presser plate contemplated by the 301 patent has "a leading surface presented towards the cylinder at an acute angle to the radius of the cylinder" (emphasis added). But that definition, if such it is, does not require that "leading surface" to be at the forward end of the presser plate. There may indeed be more than one "leading surface" on the presser plate. The front edge of the pre-tensioner B might be described as a "leading surface"; and, although there was no evidence or discussion about the precise angle of this surface in relation to the radius of the cylinder, it may well be that it was an acute angle. Be that as it may, section D has a leading edge fulfilling the requirements of the definition; and therefore the placer as a whole has "a leading surface" presented towards the cylinder at an appropriate angle to the radius of the cylinder.
 Looked at more generally, and taking a "purposive" approach to the construction of Claim 1, it seems to me that the reference to the presser plate rotating around the cylinder is intended as a reference to a plate, of whatever precise length, width or shape, which rotates around the cylinder for the purpose of pressing the warp yarn inwards so that it gets caught by the hook of the descending needle. The description "presser plate", on this approach, encompasses not only the presser plate in the 301 Patent but also its equivalent in the 198 Patent and the whole of the defenders' placer.
 Accordingly, I reject the defenders' argument that their placer comprises three separate elements, only the central one of which is the equivalent of the "presser plate" in the 301 Patent. It follows from this that, when being deflected by the pre- and post-tensioner parts of the placer, the yarn is still being deflected by the presser plate. Put another way, the pre- and post-tensioners are part of the presser plate and are not "other thread guidance or deflection means".
 Subject to a separate consideration of the operation of the weft feeder, that conclusion must lead to a finding of infringement against the defenders.
 The second argument assumes, contrary to the above, that only the central part of the Turbo 98 placer corresponds to the "presser plate" in the 301 Patent; and that the pre-tensioner and post-tensioner are both to be regarded as separate and distinct features of the Turbo 98 machine. The assumption also has to be made, which was in any event shown by the evidence and is accepted by the pursuers, that the warp yarn comes into contact with the pre-tensioner and post-tensioner and, when in contact with them, is deflected from the straight.
 On this basis, Mr Moynihan argues that the Turbo 98 machine is not a machine in which the arrangement is such that, when not deflected by the presser plate, the warp thread passes in a straight line to the top of the needle trick "without passing through or around any other thread guidance or deflection means"; and therefore does not infringe Claim 1 of the patent.
 The starting point is to identify what is meant by the expression "any other thread guidance or deflection means" at the end of Claim 1. Mr Moynihan emphasised the word "other". He submitted that the presser plate referred to in Claim 1 is a thread guidance or deflection means, in the sense that it is a means of guiding or deflecting the thread. So also, he submitted, were the pre-tensioner and the post-tensioner. They also guided or deflected the thread. They were therefore "other" "thread guidance or deflection means" around which the thread passed. I do not accept this argument. I consider that it gives insufficient weight to the word "means". In once sense, it is, of course, true to say, as Mr. Moynihan argued, that anything which, by design, causes a deflection in the thread is a "means" of deflecting the thread. But such an interpretation is, to my mind, unduly literal and wholly circular; and does not take account of the inventiveness claimed in Claim 1. The purpose of the machine, as there explained, is for the presser plate to be the "means" of deflecting each warp thread across the needle so that it is caught by the hook of the needle as it descends. The word "means" is used in this sense near the beginning of Claim 1. I would link the word "means" in the final part of Claim 1 to its meaning earlier. It is referring to a means of deflecting the warp thread into the path of the open hook of the needle as it descends. Read in this way the negative restriction in the last part of Claim 1 is not to be read as brought into play whenever the thread passes around something which causes it to deflect. Rather, it refers to the thread passing around something as a means of deflecting it into the path of the needle for the purpose of the knitting operation. "Guidance or deflection means", as that expression is used in Claim 1, are the "means" by which the thread is guided or deflected into the correct position for the purpose of the knitting operation.
 Accordingly, I do not accept Mr Moynihan's submission that, on its literal construction, or primary meaning, the negative restriction at the end of Claim 1 is engaged simply because the warp thread passes around some part other than the presser plate which causes it to deflect. I have already indicated my finding, on the evidence before me, that the pre- and post-tensioners play no useful part in the knitting operation. On that basis, one never gets to having to answer the Improver questions. The Turbo 98 machine is not a variant and the case for infringement is made out.
 On the assumption, however, contrary to that view, that, on the literal wording or primary meaning of the last few lines of Claim 1, the negative restriction is indeed engaged whenever the thread is deflected by some part of the knitting machine other than the presser plate, the Improver questions would arise for consideration. It is plain from what I have already said that I would answer questions 1 and 2 No and Yes respectively. As regards question 3, I have to ask myself whether a reader skilled in the art would have understood from the language of the claim that the patentee intended that strict compliance with this meaning was an essential requirement of the invention. To my mind, the answer to that question is plainly: No. I cannot see any basis upon which the reader skilled in the art would so have understood the patentee's intentions. The inventive idea underlying Claim 1 of the patent was the removal of the need for the thread to be held, guided or deflected by anything other than the presser plate in order for the knitting process to be effective. It removed the need, for example, for some device such as the staple in the 198 Patent. The machine disclosed in the 301 Patent can knit without any other guidance or deflection device other than the presser plate. I can think of no plausible reason why the patentee would have intended so to restrict his claim that any deflection otherwise than by the presser plate would take the variant outside the claim in the patent. I asked Mr Moynihan what the position would be if, for example, the defenders' machine incorporated a purely decorative brass plate, which happened to cause the warp thread to deflect as it went round. At one point he seemed to accept that the logic of his submission pointed to such a brass plate being a "deflection means" within the meaning of the last part of Claim 1 so as to take the variant outside the claim. He submitted that there was no infringement if the defender, using the same presser plate as the pursuers, added to their machine separate items that caused a deflection in the yarn, albeit to no purpose, and even if the addition of such items was detrimental to the efficient operation of the machine. I cannot accept that submission. It follows from this, also, that even if I had found that the thread was defected by the latch guard on the Turbo 98 machine, I would have answered the third Improver question in respect of the latch guard in the same way as for the pre- and post-tensioners.
 A separate point arose on the weft feeder. The patent does not discuss the operation of the weft feeder, simply saying that its operation is known "per se" and forms no part of the invention. The weft feeder operates in such a way that it threads the weft through the tubular net during the knitting process. The weft feeder rotates around the cylinder and inevitably causes a deflection of the warp yarn as it feeds the weft through the yarn. Mr Moynihan submitted that this was another "deflection means" within the meaning of the last part of Claim 1. I reject this argument for the same reasons as before. The weft feeder has nothing at all to do with the essential knitting operation with which the presser plate is concerned. The weft feeder is therefore not a means of deflecting the warp yarn for the purpose of engaging the warp yarn in the hook of the needle at the appropriate moment. Even if it were, I would answer the third Improver question in the same way in respect of it also.
 For these reasons I find against the defenders on the arguments raised before me on this preliminary proof before answer. It was agreed that, if this was my decision, I should simply repel the defenders' third plea-in-law. This leaves outstanding the defenders' fourth plea-in-law, which raises the question of whether the defenders can establish a defence based upon Section 64 of the Patent Act 1977.