KENNEDY, District Judge.

This is an action under the Declaratory Judgments Act, Jud.Code, § 274d, 28 U.S.C. A. § 400. The plaintiff is a New Jersey Corporation which manufactures and sells a paper feeding machine as part of a press unit sold under the trade name and trade mark “Kelly Clipper.”

There is a real controversy concerning the validity of three patents owned and controlled by the defendants, the plaintiff claiming that these patents are either invalid, or not infringed by its machine.

Defendant Dexter Folder Company is the owner of the right, title and interest in one of the patents involved here, namely, the Haupt patent.1 It also owns the Hallstream *713patent.2 It is a non-exclusive licensee under the third patent in suit, namely, the Backhouse patent.3

Defendant Harris-Seybold-Potter Company is a non-exclusive licensee under the Haupt, Backhouse and Hallstream patents and, in order to secure jurisdiction, is specifically authorized by power of attorney to act for defendant H. T. Backhouse who is, for purposes of this suit, the owner of the Backhouse patent.

The defendants in this suit counterclaim, charging the plaintiff with infringement.

The discussion that is to follow will be clearer if I say at this point that the controversy concerns itself with the automatic feeding of sheets of paper from a pile to a printing machine in lapped or streamed relationship (“stream feeding”). On this point defendants admit that the Haupt patent dominates both Backhouse and Hall-stream ; Backhouse merely introduces re- ¡ finements, and the Hallstream patent involves only apparatus claims. For this reason discussion of the principal issue in the case will center very largely on the Haupt patent. By stipulation this issue is restricted solely, so far as the validity of the Haupt patent is concerned, to Claim 1 of that patent.

The first question that must be determined is whether Claim 1 of the Haupt patent discloses a novel method of feeding sheets to a printing press. Plaintiff urges that neither Haupt Claim 1 nor Backhouse Claim 52 discloses subjects of a patentable nature, and that neither defines methods which amount to patentable inventions. It urges also that Hallstream fails to define a mechanism which constitutes a patentable invention. Plaintiff denies that there is any novelty about the method or means which Haupt and Backhouse do disclose, and says that its machine, the “Kelly Clipper,” does not infringe on any of the three patents in suit.

The Background of the Controversy. Rightly to understand the controversy which I have thus outlined, one must consider some of the problems which confronted the art of feeding sheets to a press prior to the year 1935, which marks the date of the first commercial introduction of the Haupt (Marathon) and other “stream feeders.”

There had been in vogue originally in the printing industry a system of feeding sheets to a press which I shall call serial feeding (frequently referred to in this record by witnesses as “sheet-by-sheet” feeding). A press unit actually consists of three elements: (1) a feeder, (2) a conveyor, and (3) a press. It is the feeder element which looms large in this law suit, but the interaction of the feeder and the conveyor and the press are not to be overlooked.

The sheets of paper to be printed upon are stacked in a pile or a fanned-out "bank,” from which they are separated, then removed from the pile and inserted in nip-rolls. It can be said that the function of the feeder ceases, at least in one sense of the word, when the forward edge of the sheet enters the nip-rolls-. From the latter the sheet is carried down the conveyor on tapes until its forward edge has reached the “registration guides.” These guides straighten the sheet both at the forward and the side edge so that each sheet enters the printing means (whether cylinder or flat bed) in correct alignment.

The importance of correct registration (alignment) should be clear. If the sheet is not in proper alignment as it enters the press, then manifestly the whole operation is futile, particularly in color printing. Where several colors are to be applied, it is obvious that the sheet must be in exactly the right position if it is not to be spoiled.4

*714What is the connection between the feeding means or method and registration at the printing means ? Clearly, the faster the sheet goes down the conveyor as it approaches the registration guides the less the likelihood will be that the sheet will be correctly registered. In almost every instance the guides consist of claws which seize the side and front edges of the sheet and bring them into proper alignment. Naturally, if a sheet of paper, particularly a thin one, is traveling down the conveyor at high velocity it will, when it strikes the guides, tend to buckle or bend, and poor registration will be inevitable. In fact, on most machines there is a device which stops the whole press when a sheet to be printed is not in correct alignment. And so it is important that the sheets to be printed should travel down the conveyor as slowly as possible. But under the serial feeding system (sheet-by-sheet) if the surface speed of the conveyor is slow the rate of printing will be slow because the speed of the printing means must be synchronized with that of the conveyor. In other words, there is a penalty for slow registration, namely, inability to utilize the maximum speed of the printing means.

I now return to a consideration of how this problem could be solved, or the difficulty at least lessened, by the operation of the feeder itself! I have said that the sheet to be printed is placed upon or forwarded to the conveyor by nip-rolls, which are near the forward end of the sheet supply and that is is impossible with serial feeding to bring about a slow progression of the sheets down the conveyor to the registration guides, unless the printing means is also slowed down. But many years before the issuance of the patents in suit it was discovered that if the sheets, instead of being fed to the nip-rolls in a one-by-one serial relationship, were, on the contrary, fed in an underlapped relationship or stream, they would travel • down the conveyor at the desired slow speed, and yet it would be unnecessary to slow down the printing means. The importance of the underlapped, rather than overlapped, relationship lies in the fact that the sheet to be processed must not only be free at its forward portion for some distance of its length, but also unencumbered by other sheets resting upon it, i. e., the sheet first to be seized by the registration guides must be the top sheet. This is what has come to be known as stream feeding, and during the years numerous attempts were made to develop a structure which would produce this result efficiently. Artisans everywhere seem at first to have accepted the principle that the sheet to be fed into the nip-rolls could be grasped at the front edge and then dragged off the top of the pile. Very often a stream of air was applied to the front edge of the pile near the top; this achieved much the same result as blowing upon the pages of a book to separate them. Some mechanical or pneumatic means was used to pull the front edge of the sheet into the nip-rolls. But, under this type of structure, stream feeding was found impossible, or nearly so, because the device which engaged the sheet at its front edge would, under any conceivable arrangement, not be able to seize the front edge of the second sheet until the first (top) sheet had cleared completely and was on its way down the conveyor. This would be serial, not stream feeding.

It was soon realized that this difficulty could be eliminated by a device which grasps the sheet, not at the front edge but at the rear edge. Under such a method the sheet is forwarded only through part of its length to the nip-rolls and then the next succeeding sheet can be grasped at its rear edge and forwarded underneath the preceding sheet so as to form a fanned-out stream in an underlapped relationship.

Under such a method, however, another difficulty would inevitably be produced if the sheets being operated upon were large and flimsy. It is obviously difficult to push *715them from the rear edge to the nip-rolls and still control them in such wise as to deliver each sheet with its front edge parallel to the rollers which grasp it. There is an added difficulty that a considerable amount of friction may exist between the sheet being operated upon and the rest of the pile. Furthermore, if the gripping (pushing) means at the rear edge are not exactly synchronized with each other the sheet will tend to slue around and enter the nip-rolls in a position which will forbid smooth conveyance.

Up to this point I have used vague terms for the means by which the sheet is pushed by its rear edge, or, in the early stages, pulled by its front edge. I have said that sometimes the devices were mechanical, sometimes pneumatic. In the very early forms, friction rollers made of some substance like rubber were used. These ran across the top surface of the sheet, and, if they were new, and there was no very large amount of friction, the sheets would either be pushed or pulled off the pile smoothly. But the rollers tended to become glazed in operation, and experimenters tried other mechanical means, such as fingers, which gripped the sheet at its edge and either pushed it or pulled it into the nip-rolls. Not uncommonly, and for some time prior to 1935, pneumatic grippers or suction cups were caused to descend on the top surface of the sheet and the sheet was lifted by the creation of a vacuum and then either pushed or pulled into the nip-rolls by a forward motion of the suction cups. Where these gripping devices were used at the rear edge several experimenters had also thought it advisable to impart a slight backward motion to the sheet before it was shoved forward. Obviously, this had a tendency to disengage the top sheet from the rest of the pile, to reduce friction, and thereby to bring about a smoother operation.

To summarize, there was, prior to the era of efficient stream feeding, a noticeable trend in the art toward the realization that four elements entered into the successful feeding of sheets in a stream: (1) The use of grippers, preferably suction cups, at the rear edge, (2) the injection of a blast of air at the rear edge of the sheet, (3) the imparting of a slight backward motion to the sheet before it was forwarded, and (4) the moving of each sheet forward through part only of its length and underneath the one which preceded it.5

I do not believe that any counsel in this case claims or believes that any one of these four elements in the method of feeding sheets to a press was novel at the time of the earliest application for any patent in suit. The most that can be said for the claim of the defendants is that the combination of all these features constitutes invention in Haupt because he, for the first time, discloses that smooth forward motion is produced more by the propellant effect of the air blast directed underneath the sheet at its rear edge than it is by the suction cups (or other forwarding means), also applied at the rear edge.

It will be observed that I have made specific reference to the “propellant effect of the air blasts.” Obviously the separation of sheets of paper by blowing upon them was not novel at the time of the issuance of the patents in suit. But the defendants in this case claim that Haupt taught something new to the art, namely, that the air blast directed at the rear of the pile performs two distinct functions: (1) Separation of the sheets, and (2) forward propulsion of the sheets into the nip-rolls. In fact, tlie defendants argue in substance that the suction cups, which almost invariably are referred to in the art as the “forwarding means,” are actually not that at all, but merely devices to control the sheet so that it rides smoothly on the air current, which is the real forwarding means, or at least the major element in the forwarding operation. Defendants say that this was not known in *716the art before Haupt, that he discovered and disclosed it, and that this discovery, even in combination with known elements, constitutes invention.

All of this explains why in the trial record of this cause there is much debate on two questions: (1) Whether the sheet is forwarded to any great extent by the suction cups at its rear edge, or whether the air blast does all or most of it, and (2) what various patentees prior to Haupt mean when they use expressions like “film of air,” “blast of air,” or “current of air.”

This brings me to what I consider to be the issues in the cause, namely: (1) Does Haupt in point of fact disclose that in stream feeding the air blast is the principal, if not the only forwarding means? (2) If Haupt does make this disclosure, is it invention? And (3) does the accused machine infringe upon the patents in suit, assuming them, or any of them, to be valid?

The Patents in Suit, and Particularly the Haupt Patent.

At the risk of wearisome repetition, I say again that the defendants’ claim is that Haupt first disclosed the propellant effect of air blast in stream feeding. Therefore, the defendants admit that Haupt dominates Backhouse and Hallstream in this respect— the principal point to be considered in the cause.

Perhaps the main accusation leveled against Haupt by the plaintiff is that he really does not, under the statute (R.S. 4888, 35 U.S.C.A. § 33), disclose the importance of the air blast as a forwarding means in stream feeding.

The Haupt patent is very brief (pi. ex. 1). It begins with an admission that separation of sheets at the rear edge by some mechanical means, aided by an air blast, is already known (Haupt p. 1, lines 10-14). It is then asserted that the difficulties standing in the way of accurate registration can be removed by stream feeding (Haupt p. 1, lines 30-36). Reference is then made to the drawings, which are five in number.

In the specifications (Haupt p. 2, lines 10-34), Haupt describes a means for interrupting feeding of the paper which is not important here. This reduces the specifications, so far as the issues at bar are concerned, to 49 lines, and of these it can fairly be said that only five are important, namely, the following sentence (Haupt p. 1, lines 68-72):

“The rod then moves into the position shown in Fig. 2 and lifts the edge of the sheet into contact with feeding claws 7, the sheet being separated from the pile by a blast of air from nozzles 8.”

Testimony about the construction of this sentence recurs over and over in the record. However, the claim of the defendants is not based solely or even principally on this sentence in the specifications. I think it is fair to say that in construing the patent the defendants, as I believe they freely admit, have paid little attention to the specifications.

The argument of the defendants is actually based upon their interpretation of Claim 1 (Haupt p. 2, lines 35-48) read in the light of one diagram, Fig. II. It is easy to see why defendants are compelled to take this position if they are to succeed in their contention that Haupt discloses the importance of the air blast as a propellant means in stream feeding. In the major portion of the specifications (Haupt p. 1, lines 60-98) the patent discloses as clearly as any language can say it that Haupt believed that the sheet to be fed could be pushed forward by mechanical claws engaging its rear edge, after it had merely been separated from the rest of the pile by a blast of air. It should be unnecessary to support this statement by any extended reference to the patent itself. The patent calls these claws “feeding claws,” and, describing their operations says, to paraphrase: - “The rod * * * lifts the edge of the sheet into contact with feeding claws * * . The feeding claws * * * push the front edge of the sheet * * * under rollers * * *.” (Haupt p. 1, lines 67-78). What Haupt is describing could not be clearer. He says that a suction member (Fig. I, 4) comes down on the rear edge of the sheet to be fed, and lifts it while it is being separated from the rest of the pile by an air blast (Fig. I, 8). The suction member (Fig. I, 4) then lifts the sheet into contact with the “feeding claws” (Fig. I, 7) which engage the sheet at the rear edge and push the front edge into engagement with nip-*717rolls (9). At this portion of the patent the air blast is mentioned only very incidentally, and then as a separating means and not as means of propulsion. This would seem to put entirely out of consideration any disclosure in the Haupt patent such as the defendants claim.

However, the defendants say that when Haupt talks about feeding claws pushing the rear edge of the sheet into engagement with the nip-rolls (Haupt p. 1, lines 60-98), and when he says, for example, “each sheet is carried by the claws 7 under the rear edge of the next preceding one * * * ” (Haupt p. 1, lines 88, 89), he is confining his discussion to the use of his device for a fanned-out “bank” as disclosed in Fig. I and that he is not talking at all about the use of the device for feeding sheets from a flat pile. For that purpose, they say, he discloses a different method.

Defendants actually rely upon three peculiarities in the patent to sustain their contention that two methods are disclosed. What are the peculiarities — the internal evidence — upon which this distinction (i. e. between Fig. I and Fig. II) is based?

(a) In the Specifications.

The defendants say Haupt makes this distinction in the specifications because he prefaces the major part of these, which weigh against thqir contention (Haupt p. 1, lines 60-98), by the qualifying words “in the arrangement according to Fig. 1” (Haupt p. 1, line 61; emphasis mine). After he has completed this discussion they say he turns to flat pile feeding, and they point to the fact that he prefaces the second portion of his specifications by the qualifying words “In the case of a flat pile as shown in Fig. 2” (Haupt p. 1, line 99; emphasis mine).

This bit of internal evidence is open to very obvious objection, both on the positive side and on the negative side. In the portion of the specifications which deal with flat pile feeding as a separate matter (Haupt p. 1, lines 99, 100; p. 2, lines 1-9) the inventor contributes nothing new to the former specifications, except a description of a device at the forward edge of the pile in the form of an air blast striking the undersurface of the sheet merely to position it at the nip-rolls. On the negative side Haupt, in the portion of the specifications which deal with flat pile feeding, reference to which I have already given, says nothing at all about a means of forward propulsion different from that which he has already talked about (the feeding claws). In other words, standing by itself, the claimed distinction in the specifications between feeding from a fanned-out “bank” (Fig. I), and feeding from a flat pile (Fig. II) is probably an empty one, for anyone reading the patent would naturally imply that Haupt was teaching concerning flat piles the use of the same forwarding means that he had previously discussed for fanned-out “banks,” and was merely adding the admonition that operation on flat piles would be improved by an air blast also at the forward edge, which would position the sheet better. The reason why he probably meant that, and nothing more, is obvious if one compares the drawing for the fanned-out pile (Fig. I) with the drawing for the flat pile (Fig. II). In the former the sheets approach the roller at an angle inclined downward and would naturally tend to drift into the'bite of the rollers (9). In the latter (Fig. II) the sheets being horizontal on the pile to start with, would probably need to be blown over the lower nip-roll edge, which, in the drawing, is placed at almost the same level as the sheet itself, if not slightly higher.

And so it is clear that if defendants’ claim (that Haupt disclosed an air blast at the rear edge as the chief propelling means) is based solely on the theory that Haupt was trying to describe two different devices, it would, so far as the specifications alone are concerned, be very weak, if not totally unsupported.

(b) Claim 1.

But the defendants go further. They say that Haupt, in the only claim in suit here (Claim 1), defines a procedure for feeding the sheets into the nip-rolls separate and apart from that involving the use of mechanical feeding claws at the rear edge, and that the only fair inference to be taken is he is claiming that the sheets are forwarded by the air blast. And so I *718turn to the examination of Claim 1' (Haupt p. 2, lines 35-48).

It is perfectly obvious from a cursory reading of the claim itself that Haupt never once specifies that the air blast is the forwarding means. The claim has four elements: (1) Lifting the sheets by suction from the pile at the rear edge, (2) carrying each sheet forwards horizontally under the preceding sheet into contact with the feeding elements to form a thin continuous layer of overlapping sheets, (3) advancing such layer at each operation of the machine through part only of the distance between the machine and the feeding device, and (4) adding one sheet to the layer for each sheet taken away therefrom by the printing machine. I have actually quoted the claim in detail at this point, eliminating only conjunctions and some preliminary matter. Obviously, the elements 1, 3 and 4 in my analysis of the claim are not important for the business at hand. The first element, lifting the sheets by suction at the rear edge is not new, as Haupt himself admits (Haupt p. 1, lines 1-4). The third and fourth elements of the claim describe stream feeding, which, as I have already hinted, can hardly be considered novel by itself. And so defendants’ contention concerning Haupt’s Claim 1 must turn on the meaning of the words in which he sets forth the second element, namely, “carrying each sheet forwards horizontally under the preceding sheet into contact with the feeding elements.”

In support of their proposed interpretation the defendants say that the significance of this language is brought home if consideration is paid (1) to the fact that Haupt uses the word “carrying” and (2) that he uses the word “horizontally.” The defendants say that Haupt deliberately omitted any reference to feeding claws pushing the sheet forward, that he chose rather to use the word "carrying,” which would be inappropriate to describe that operation, and therefore Haupt intended (by implication) to disclose the air blast,"and nothing else, as the means of propulsion. Defendants say that if any further evidence were needed it would lie in the fact that 'Haupt uses the word “horizontally,” which would probably be impossible of accomplishment where a thin sheet is pushed from its rear edge by mechanical claws as the sole forwarding means. Finally they say that Haupt is talking of a means of feeding sheets from a pile (Haupt p. 2, line 37), and that this correlates the claim with the distinction already" detected throughout the specifications and previously discussed, and also with the .distinction between the diagram Fig. I and the diagram Fig. II, to be further noticed.

It should be obvious what the position of the plaintiff is concerning Haupt’s Claim 1. It says that having previously described mechanical claws or an equivalent as the forwarding means in his method Haupt did not choose to repeat that in the claim, but rather implied it.

It does seem that the defendants are basing their'contention here, as in their construction of the specifications, more upon what is not found in the language of the patent than what is found there. For the moment, I leave this and proceed to a consideration of the diagrams.

As I have tried to indicate up to this point, defendants, in their construction of the Haupt patent (basing their claim upon the distinction supposedly made by Haupt in the specifications between feeding from fanned-out “banks” on the one hand, and feeding from a flat pile on the other), urge that Fig. II is intended to depict a device essentially dissimilar from that shown in Fig. I. A very large amount of the testimony at the trial concerned itself with the meaning of the former diagram. I will try to make these passages of the record better understandable by a summary' of the differences of opinion that arose with respect to it (Fig. II).

One controversy relates to the action and positioning of the suction cup (Fig. II, 4). The defendants point to language in the specifications (Haupt p. 1, lines 64 — 70) where the suction cup (Fig. II, 4) is called a “suction rod,” language that I now paraphrase :

“A suction rod provided with a plurality of nozzles 4 is operated in known manner and applied to the rear edge of the top sheet * * *. The rod then moves into the *719position shown in Fig. 2 and lifts the edge of the sheet into contact with the feeding claws 7 * *

The defendants, for this purpose construing Fig. I and Fig. II together, and as representing progressive stages of the same operation, say this language means that the suction rod or member has an upward and forward motion. They urge that this must be true, because otherwise the suction member could never bring the rear edge of the sheet into contact with the feeding claws. Plaintiff, on the other hand, asserts that the suction member has merely an upward motion, and no forward motion whatever, its argument being based upon the fact that the nozzle (8) is a fixed member of the assembly, and that the suction member (4) is disclosed in the same position relative to this fixed member in Fig. II as it is in Fig. I. In answer to this, the defendants say that this is accounted for on the basis that the nozzle (8) is not a fixed member, but moves forwardly with the suction rod. They point to the fact that while the suction member has not changed its position in Fig. II relative to the nozzle, it has changed its position relative to the rear edge of the pile, which in Fig. I is indicated by the register (3) and in Fig. II is plainly displayed.

The fact is that on the basis of such a sketchy and obscure drawing it is extremely difficult to say what the fact is with any degree of assurance. If Fig. I is to be considered as depicting a device wholly distinct from that which is disclosed in Fig. II, as the defendants reason in other places, then of course we are left with nothing at all to guide us concerning the true action of the suction rod. If on the other hand the figures are to be read together for this particular purpose, then to reach the result contended for the defendants, we must assume that the air nozzle (8) has a motion forward and backward, although nowhere else in the patent is this even suggested.

But I believe that a determination concerning the exact motion of the suction rod (4) in the Haupt patent is not strictly necessary. I say this because it is defendants’ contention that the main forwarding means is the air blast, and not the operation of the suction rod. Indeed, concerning Fig. II defendants’ witnesses plainly state that the “feeding claws” themselves perform no really essential function in forwarding the sheet, but serve only to steady it. If this be so, namely, that the air blast is intended to be the principal forwarding means then, as I have said, whether or not the suction rod also contributes a forwarding effect is not really vital.

A second controversy, arising (at least as I think) out of the obscurity of Fig. II of the Haupt patent, is whether in its present form and if defendants’ claim has substance, that drawing discloses an operable structure. It will be observed that in the drawing itself (Haupt Fig. II) the suction rod (4) is shown elevated completely above the sheet, and not in contact with it. It therefore exercises no control. Moreover, the forward end of the sheet while it is within the range of the bite of the nip-rolls (9) is not shown physically in contact with the rolls. These nip-rolls, therefore, do not control the sheet And yet the defendants put the greatest stress on the fact that in this drawing the sheet is shown wholly lifted from the pile, in approximately a horizontal position (Cf. Haupt, p. 2, line 39). The defendants attribute the utmost importance to this feature of the diagram, because they say it, read in conjunction with Claim 1, means that the air, and the air alone, is supporting and forwarding the sheet. Plaintiff on the other hand, and with some considerable color of reason, says that a sheet subjected to an air blast strong enough to hold it separate from the pile, as shown in Fig. II of Haupt, would be blown out of the grasp out of the claws (7) and would be completely out of control.6

It may seem curious that I pay so much attention to a rough diagram or drawing in *720the patent.7 I do so only because the defendants do, and if they in interpreting that drawing by their exhibits (defs. exs. E.E. and G.G.) felt themselves required to make the modifications to which I advert, then obviously that fact has a bearing upon the weight which must be given to the diagram Fig. II.

■ But there is an additional reason for touching upon this matter at this point, as will be seen more clearly later. One of the principal arguments of the defendants in the case at bar is that a vital element of the Haupt patent, and of the Backhouse patent, too, is that the suction member (roughly equivalent to the suction rod (4) in the Haupt patent) is not so important in forwarding the sheet toward the bite of the nip-rolls as it is in steadying the sheet and controlling it while the air blast advances it. Witnesses for the defendants went so far as to say that where, for example, Backhouse in his patents refers to a set of “forwarding suckers,” he is really inaccurate in his language, and means "control” suckers. And so the importance of this controversy about the interpretation of Fig. II of Haupt may become a little clearer. It is the only drawing, say the defendants, which discloses Haupt’s discovery to the extent that feeding from a flat pile is concerned. The clear space between the top sheet and the pile, they say, means that the sheet is being forwarded horizontally into the nip-rolls on a blast of air. The natural inquiry of the plaintiff is: why, if this is Haupt’s meaning, does he omit to disclose in that same drawing a very essential feature of what is claimed to be his method, namely, control of the sheet while the air blast is sending it forward? Obviously, says'the plaintiff, this control is not supplied by the suction rod: Haupt shows it out of contact with the sheet. Obviously, plaintiff urges, this control is not supplied by the nip-rolls: they are shown in an open position. And to cap all of this, the defendants, so far as Haupt’s Fig. II is concerned, have beeni compelled to disclaim the feeding claws (7) as a forwarding means, and even to minimize greatly its importance as a control means. I would say that the defendants’ principal answer to all this is that in his Fig. II Haupt has chosen to display a sheet during that transitory instant (augenblick) when the sheet is free of the control of the sucker rod (4) and just about to be gripped by the nip-rolls (9). This would be the worst possible instant to select, if Haupt means what defendants say he does.

(d) Summary.

I have previously dealt with the defendants’ analysis of (a) The Specifications, (b) Claim 1, (c) The Diagram Fig. II, all in the Haupt patent, in that order. For purposes of summary I now reverse that order.

It will be at once seen that the difficulty of the defendants’ position concerning the diagram Fig. II is that Haupt, for some esoteric reason, selects for this important drawing the very transitory instant when, if his drawing is literally read, the sheet would undoubtedly be out of control, and an operable structure would not result.

Turning to Claim 1: the weakness of the defendants’ position is that again, for some mysterious reason, Haupt, wishing to disclose that the sheet was forwarded by a blast of air, chose not to mention this specifically, but instead used a vague word like “carrying” without disclosing the precise carrying means (Haupt p. 2, line 39).

As for the specifications: The weakness of the defendants’ position is that they are compelled to say that Haupt was actually speaking of two different devices, one applicable to fanned-out “banks,” and the other to flat piles. Haupt’s specifications are complete and clear concerning the former, where he says plainly that the *721feeding claws (7) perform the forwarding function. But for flat pile feeding, the defendants say we must take the implication that Haupt was discarding the use of the forwarding means which he had just spoken about and, without mentioning it specifically, was teaching those skilled in the art to forward the sheet by means of a blast of air.

From all this, the only conclusion that it is possible for me to reach is that Haupt’s patent complies with neither one of the commands found in the statute, R.S. § 4888, 35 U.S.C.A. § 33, so far as the present controversy is concerned. He does not disclose a method of stream feeding under, which the air blast is the principal forwarding means; neither is his Claim 1 so sufficiently distinct and definite that it makes known features of Haupt’s method which may not be safely used without a license. General Electric Co. v. Wabash Appliance Corporation, 1938, 304 U.S. 364, 58 S.Ct. 899, 82 L.Ed. 1402. It is urged on behalf of the defendants that plaintiff is disabled to argue that Haupt’s disclosure is inadequate because it did not support this contention by the testimony of a witness skilled in the art, but relied solely upon evidence given by a patent expert. There is language in the books which seems to sustain defendants’ argument. Walker, Deller’s Ed., vol. 3, sec. 728, p. 2035, and cases cited. I can well understand how a trier of the facts might take this position when the patent involved related, for example, to a chemical composition or process, or to an improvement in a loom. But still it seems to me that the peculiar qualifications of any witness merely go to the weight to be given his testimony, and where, as here, the issue of disclosure must be resolved by the construction of plain English uncomplicated by technical terms, then the defense of inadequate disclosure can be dealt with even without the evidence of a skilled artisan.

If I am right in the conclusion that neither Haupt nor Backhouse clearly discloses a method of stream feeding under which the jet of air plays the main part in forwarding the sheet, it is unnecessary for me to consider the questions of novelty, invention, or infringement. However, it may be of some help to a court of review if I do consider these matters, even at the risk of expanding a decision already too long and tedious. I say this because counsel in this case, in their zeal to make sure that I understood the issues, were kind enough to afford me the opportunity to watch in operation, as well as in moving pictures, not only the accused machine but structures claimed to be embodiments of the Backhouse patent. This inspection convinced me about certain matters which I think I ought to make available to the reviewing court, for what my observations are worth, even though they touch upon issues unnecessary to decide.

There is not the slightest doubt in my judgment that as stream feeding is now practiced the air blast is in fact the principal forwarding means in all of the machines that I saw.- The suction cups at the rear edge of the sheet being forwarded do little more than control that sheet while the air blast is administered. I believe that the so-called forwarding suckers, particularly in the embodiment of Backhouse, merely position the sheet which, upon release, floats into the nip-rolls on the air jet. Moreover, as each sheet advances, the strong forwardly and upwardly current of air binds the sheet operated upon to its predecessor so that a cohesive stream is. formed.

Neither is there any doubt in my mind that the use of the air blast, as practiced since about 1935, is the principal factor in the somewhat sudden commercial success of stream feeding.

It is not too much for me to say that the defendants base their argument in this case very largely upon these two facts (i. e., that the air blast plays the principal part, and that stream feeding became commercially successful immediately after this fact became realized).

If all this is so, the defendants argue, how can one deny that Haupt (1933), assuming that I am wrong on the issue of disclosure, had discovered a new and useful method of stream feeding? The defendants point to the Eibel case. Eibel Process Co. v. Minnesota & Ontario Paper Co., *7221923, 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523. There the Supreme Court upheld the validity of a patent which improved the operation of a w-idely-used paper machine merely by raising the breast-roll and of the moving wire screen. The defendants say that this case is a perfect pattern for the ease at bar on the issues of novelty and invention here. (Throughout, in what I say hereafter I am assuming that I am wrong in my conclusions about what Haupt discloses, and that the questions of novelty and invention are here for decision.)

In the Eibel case, it seems to me, the Supreme Court went very far in the direction of liberality.8 But I believe also that there is a substantial difference between the case at bar and the Eibel case, for nothing in the prior art seems to have suggested Eibel’s solution of the particular problem with which he dealt, whereas, the art prior to Haupt in this case certainly does suggest that air was, at least in part, a means of propulsion of sheets to be forwarded to a printing press. This is certainly true if I give to the prior art patents in this case, as well as to Haupt and to Backhouse, the liberal construction that was given the patent in the Eibel case.

Vickery (British No. 19,335, 1904, pi. ex. 9) does not talk about stream feeding in terms, but he does describe a method of feeding sheets of paper to printing, ruling and like machines. To me, at least, he places considerable emphasis upon the forwarding role played by the air blast (see for example Vickery, p. 3, lines 38-43, pi. ex. 9). Smythe (Ü.S. 779, 975, applied for November 14, 1901, issued January 10, 1905, pi. ex. 10) described a sheet-feeding machine. This inventor not only clearly appreciated that the air blast performed a forwarding function, but also described that function in terms much clearer than those employed by Haupt (Smythe, pi. ex., 10, p. 2, lines 57-65) : “the sheet, by means of the air blast, is free from contact or adhesion to the sheet below, the attractive force being destroyed by the action of the air blast, and rides upon the air in a way which will permit it to be pushed forward by the- return forward movement of the grippers, until the forward edge of the sheet engages the feed rolls 9, which then take it and carry it off the pile.”) Defendants’ criticism of Smythe’s patent is that he provides a film of air merely as a lubricant. But Smythe, in the passage I have quoted, says in his specifications that the sheet “rides upon the air.” And so it seems to me that even though Smythe claims not a method, but a machine, his patent teaches the importance of the air blast much more clearly than Haupt and long before the latter.

Thus, defendants, on the issues of novelty and invention, are in an untenable position. They admit that Pallister (pi. ex. 6) sought to accomplish stream feeding. And it is difficult for a trier of the facts not to conclude that air was known as a forwarding means prior to Haupt. Yet defendants say that Haupt discovered something new and useful, assuming that he disclosed a method of stream feeding in which the air jet plays the principal part-as a forwarding means. It seems to me that it is impossible to support this position, and if the questions of novelty and invention were before me I would decide them against the defendants.

It is perfectly true that it is not easy to rationalize, on the one hand, the state of the art prior to 1935, and, on the other, the sudden commercial success of the air-forwarded method of stream feeding. My own belief is that artisans dealing with press feeders from Vickery (1904) to Haupt (1933) had all realized that air plays some part in the advancement of the sheet. But their minds may have instinctively recoiled from attempting the necessary mechanical adjustments which would minimize the forwarding function of the suction cups, and would utilize to the fullest the forwarding function of the air blast. In all liklihood, they believed that an inoperable structure would result if they did this: inoperable, because, particularly with flimsy sheets, it would not be possible to control and position them if they were subjected to a heavy air blast.9 It seems *723to me quite likely that, under pressure of commercial demands, increased experimentation with structures already in use, and mechanical adjustments on them, led to the successful commercial operation of stream feeding: that invention, in the statutory sense of the word, played no part in it.

I will be criticized for making this sweeping conclusion on the ground that it is not supported by any clear evidence in the record. There was, however, evidence of one curious incident, of which both sides attempted to make capital, an incident that took place when the witness Harris, representing one of the defendants, set about acquiring a license under the Backhouse patent. I ought to say at this point that Harris was a very frank and helpful witness, and also that he has not only a technical but a practical background in the art of press feeding.

Some time in 1934 Harris, having observed the Backhouse feeder working in England, became anxious to acquire a license, and did so. When he returned from England he explained the construction of the feeder to his engineers, and on this basis they built one. But it never operated satisfactorily, as a result of which Harris became the butt of their jokes. Ultimately there arrived from England a Backhouse feeder which Harris had purchased. This machine was highly satisfactory in operation. The engineers then discovered what Harris himself did not know, even after observation of the feeder in England, namely, that it was the larger volume of air that accomplished the result. Harris admitted that the experimental model, constructed upon the basis of his description of the Backhouse feeder, would have been equally satisfactory in operation if he had merely increased the air blast. I think this incident at least suggests that the commercial success of stream feeding was produced by a mechanical adjustment that might easily have been made on structures and in methods already known and in use.

Again it may be that a court of review will not be willing to accept the views which I express on the issues of disclosure, novelty, and invention. And again it may be helpful if, on the issue of infringement, and for what it may be worth, I play the role of a witness who has seen the machines in operation. If the Haupt patent is valid, there is little doubt in my mind that the plaintiff is guilty of infringement.

Clearly, the “Kelly Clipper” which I saw in operation is a structure which performs a method of stream feeding, the sheets being removed from the pile, and forwarded in an underlapped relationship, the principal forwarding means being the air blast. The suction cup in the “Kelly Clipper” moves backward, as in Backhouse. Only one suction cup is employed, as in Hall-stream, this one cup being in part a separating means and in part a forwarding means, so that as a device the “Kelly Clipper” is practically identical with Hall-stream’s machine.

The plaintiff argues that there is a substantial difference between the “Kelly Clipper” and any possible embodiment of the patents of Haupt, Backhouse and Hall-stream because in the accused machine the sheet to be operated upon is not gripped and lifted from the pile by the suction cup but is, on the contrary, blown up toward the suction cup by the blast of air. This difference seems to me inconsequential. Plaintiff argues further that in the accused machine the backward movement of the suction cup, described in Backhouse, is primarily for the purpose of disengaging the forward end of the sheet from a hold-down clip. No doubt that is one of the functions of the suction cup in the accused machine, but it is clear that it performs another function also, namely, the separation of the sheet from the pile as in Back-house. And again it seems to me that the fact that it performs the additional function of disengagement is not important.

Plaintiff is entitled to a judgment, and to a dismissal of a counter-claim on the ground that the key patent, that of Haupt, *724does not disclose a method of stream feeding which employs the air blast as the principal means of forwarding the sheets to be operated upon.

I have filed findings of fact and conclusions of law.

This is U. S. patent 1,898,535. It was issued February 21, 1933, having been filed in the United States on January 30, 1930. The patentees are A. Haupt, H. Gumbel and J. Geisecke. It relates to a method and means of feeding paper sheets. By stipulation only Claim 1 of the Haupt patent is involved here.

This is U. S. patent 2,144,057 issued to H. Hallstream on January 17, 1939, filed in the United States on April 21, 1937. This patent relates to a paper feeding machine. By stipulation Claims 1, 2. 3, 4, 6, 9, 10 and 26 of the Hall-stream patent are involved here.

This is U. S. patent 2,108,702 issued on February 15, 1938, to H. T. Back-house; filed in the United States on August 4, 1934. This patent relates to automatic sheet feeding machinery. By stipulation only Claim 52 of the Back-house patent is involved.

Perhaps the importance of correct registration in color printing is best realized when one compares the multiple color work found in the average comic *714strip with that in the better type of magazine. In the comic strip very often the colors are either run together or do not join at the proper line, whereas in the better magazines it is difficult to notice irregularities of that nature even with a magnifying glass. This is because the well-printed sheet has been correctly registered when it entered the printing means and each of the several colored impressions is in exactly the right place.

It will be seen that the whole operation embracing these four elements is not unlike that by which the housewife straightens out a tablecloth. She grasps it with both hands at the rear edge, draws it backward a little, and then whips it up and forward until it rests squarely on the table. Her grasp at the edge is the equivalent of the suction cups. The slight backward motion is to free the cloth of any friction which may exist between the table and cloth itself, and the whipping forward motion is produced by a combination of the action of the forwarding means and the operation of the air underneath the sheet.

I call attention to the fact that, in their enlarged drawings submitted in evidence to explain Haupt’s patent, the defendants never show the sheet being operated upon in a position precisely simto that disclosed in Fig. II of the patent. Thus in their own diagrams, in the instant before the nip-rolls have grasped the front edge of the sheet they show it held by the suction rod (defs. ex. E.E., Fig. II; defs. ex. G.G., left hand column, Fig. H). On the other *720hand, after the suction rod (4) has been disconnected and at a time when the feeding daws (7) are merely touching the rear edge of the sheet, the defendants in these diagrams show the forward edge of the sheet within the bite of the nip-rolls (e.g. defs. ex. E.E., Fig. Ill; defs. ex. G.G. left band column, Fig. III).

Perhaps both the defendants’ counsel and myself are giving to patent drawings an importance which the Supreme Court says they do not deserve. Permutit Co. v. Graver Corporation, 1931, 284 U.S. 52, 60, 52 S.Ct. 53, 76 L.Ed. 163.

See In re Jennings, 1943, 133 F.2d 906, 908, 30 O.O.P.A. (Patents) 887; James Heddon’s Sons v. American Fork & Hoe Co., 6 Cir., 1945, 148 F2d 230, 233.

This also harks back to the question *723of disclosure: I think that the average artisan, looking at Fig. II of Haupt, (pi. ex. 1), would never believe that he was teaching the application of a strong blast of air to forward the sheet into the nip-rolls. He would reject this conclusion on the hypothesis that the sheet would be blown out of control.