FOXBORO CO. v. TAYLOR INSTRUMENT COMPANIES, 157 F.2d 226 (2nd Cir. 1946)


FOXBORO CO. v. TAYLOR INSTRUMENT COMPANIES.

No. 188, Docket 19997.Circuit Court of Appeals, Second Circuit.
July 25, 1946. Rehearing Denied September 12, 1946. Writ of Certiorari Denied January 6, 1947. See 67 S.Ct. 494.

Appeal from the District Court of the United States for the Western District of New York.

Patent infringement action by Foxboro Company against Taylor Instrument Companies. From a judgment, 58 F. Supp. 313, holding valid and infringed all claims but

Page 227

No. 18 of Reissue Patent 20,092, issued on September 1, 1936, to Clesson E. Mason, in substitution for Patent No. 1,897,135, issued to Mason on February 14, 1933, the defendant appeals.

Judgment reversed and complaint dismissed.

Drury W. Cooper, of New York City, D. Clyde Jones, of Rochester, N.Y., and Drury W. Cooper, Jr., of New York City, for appellant.

Charles C. Ladd and Edward G. Curtis, both of New York City (Marshall M. Holcombe, Gordon K. Lister, and Blair, Curtis
Hayward, all of New York City, of counsel), for appellee.

Before L. HAND, SWAN, and CHASE, Circuit Judges.

L. HAND, Circuit Judge.

The defendant appeals from a judgment holding valid and infringed all claims but number eighteen of Reissue Patent No. 20,092, issued on September 1, 1936, to Clesson E. Mason, the original having been issued on February 14, 1933 (No. 1,897,135), upon an application filed on September 15, 1930. The district court discussed the case at length in a published opinion (Foxboro Company v. Taylor Instrument Companies, 58 F. Supp. 313), to which we refer for an account of the general nature of the invention, of the specifications and claims, and of the contentions of the parties. Assuming that the reader will be familiar with these, we find it necessary to consider only two questions: (1) Whether Mason forfeited his patent because of his failure to disclaim certain of his claims; and (2) whether the claims are invalid, because they do not describe any patentable advance over the art, as it had developed on September 15, 1930.

The answer to the first question depends upon the effect of Mason’s disclaimer of claims two, three and five of the original patent on May 11, 1933, shortly after it issued. Claim one and claim four were retained, and claim four is precisely the same as claim two of the reissue. A different question would have arisen, had original claim four not been in the patent, and had Mason disclaimed claims two and three and then attempted to revive claim three, adding to it the new element by which original claim four was different from original claim three. Since, however, he carried original claim four into the reissue unchanged (as claim two), there never was a period during which he was not claiming the invention which it defined; and the question comes down to whether the disclaimer of original claim two and three demanded the disclaimer also of claim four. In Maytag Co. v. Hurley Machine Co., 307 U.S. 243, 59 S.Ct. 857, 83 L.Ed. 1264, the Supreme Court held that, when a patentee had disclaimed two claims which had been held invalid, but failed to disclaim a third which was not “definitely distinguishable” from the disclaimed claims, he forfeited his patent. It is true that Roberts, J., added (307 U.S. at page 245, 59 S.Ct. at page 859) that the owner of a patent, “by disclaiming those claims, has confessed that the patentee therein claimed `more than that of which he was the original or first inventor or discoverer'”; and the defendant argues that this confession must be considered a conclusive admission that the matter contained in original claims two and three was in the prior art, from which it would follow that the differentia of the original claim four from those claims must be enough to constitute invention. In the case of claim four the differentia was as follows: “Said chamber being adjustably movable as a whole and by said adjustment affecting movement and adjustment of one of said elements relative to the other.” This language can only refer to the fact that bellows, “39,” can be moved by means of the “frame, 45,” so as to adjust the nozzle to the flapper; and that, the defendant argues, would not require invention. If the language quoted from Maytag Co. v. Hurley Machine Co., supra,[1] be read literally, this argument would not be without force; but a number of courts — beginning with Judge Archbald’s decision in Manhattan General Construction Company v. Helios-Upton Co., C.C. 1905, 135 F. 785, 802 — have decided that a disclaimer should not be taken as an admission by the patentee that the disclaimed

Page 228

matter was not patentable;[2] and we do not believe that the Supreme Court intended to overrule this doctrine. The consequence of so holding would be indeed severe, for it would result, whenever the patentee made any disclaimer, that he risked his whole patent except in so far as his remaining claims were patentably distinguishable from those disclaimed. In United Chromium, Inc., v. International Silver Co., supra,[2] 60 F.2d at pages 914, 915, we stated our reasons for thinking that by disclaimer a patentee does not so concede. We have tacitly assumed for argument that the doctrine of Maytag Co. v. Hurley Machine Co., supra,[3] applies as much when a patentee voluntarily disclaims one claim, but fails to disclaim another which is not “definitely distinguishable,” as when a court has declared a claim invalid. We do not, however, mean to express any opinion on that question.

We limit our consideration of the validity of the claims to those on which the plaintiff alone relies: reissue claims two, five, ten and thirteen. Of these four, two and ten are “proportioning” claims, and claims five and thirteen are “reset” claims. All these were in the original patent in substantially the same form that they appear when reissued; and the same indeed applies to all the first sixteen claims of the reissued patent. As to these we need not consider whether Mason showed adequate excuse for any reissue whatever, because, even though he did not, the surrender of the original patent, resulting from the acceptance of the reissue, did not invalidate any claims which he carried over into the reissue. For a time this was in some doubt (Eby v. King, 158 U.S. 366, 15 S.Ct. 972, 39 L.Ed. 1018), but the amendment to § 64, Title 35, U.S.C.A., in 1928, has now set the question at rest. Schenk v. United Aircraft Corp., D.C., 43 F. Supp. 679, 686. Although that amendment uses the word, “identical,” we read this as “substantially identical,” and the first sixteen reissued claims are substantially identical with their originals. Finally, we do not think that claims two, five, ten and thirteen are too vague to be valid. They are, indeed, stated with the turgidity customary in such documents, but it is possible to identify with sufficient certainty those parts of the disclosure to which they refer. We proceed therefore to consider their validity in the light of the art.

In November, 1924, Thomas Lindsay filed “complete specifications” for a patent which issued on May 28, 1925 (1925, Brit. Pat. 234,194), for a controlling apparatus of the same general class as Mason’s. The single figure shown is for the control of the pressure in a steam pipe by means of a steam valve, operated by a pulley and weight which move it in one direction, and a piston and cylinder which move it in the opposite direction. We can see no patentable distinction between moving the valve in this way and attaching it to a diaphragm opposed by a spring; and indeed, Lindsay, who had described such a diaphragm in an earlier patent (1924 Brit.Pat. 218,370) incorporated this in his patent as an alternative (page 5, lines 37-41). The piston is moved against the weight by the pressure of a “non-compressible” fluid, introduced into the cylinder, whose waste from the cylinder allows the weight to pull the piston back. Control of the steam pressure is accomplished by tapping the main pipe by the lead, “35,” to the bellows, “34,” in a steam tight chamber. The compression of this bellows raises the “lever, 37,” against the pressure of the “spring, 38,” and at the end of this lever is the “cup, 31,” which acts as a baffle for the nozzle, “23.” So far as we can see, this is a precise analogue of Mason’s flapper and Bourdon tube, except that Lindsay’s regulating device — bellows, “34” with its spring, “38” — and his damping device — the bellows, “43” with its spring, “41” — act upon the flapper. The plaintiff sets great store on this distinction.

Page 229

The “non-compressible fluid” which enters the piston comes into the control system by means of the “pipe, 13,” which ends in the “valve, 14,” which, when it is open, leads in turn through the “pipe, 11,” to the cylinder, and when closed, permits the cylinder to waste the fluid. Branching from the “pipe, 13,” before it reaches the “valve, 14,” is the lead, “20,” which ends in the “needle valve, 18,” beyond which is the lead to one side of the “diaphragm, 16,” the other side of the diaphragm being in contact with the stem of the “valve, 14.” Beyond the “needle valve, 18,” and, opposite to the lead into the diaphragm, the “pipe, 21,” leads away and ends in the nozzle, “23,” for which the “cup, 31” acts as a flapper. If we assume the “cup” to be sealing the nozzle, the pressure in the “pipe, 13,” will be held by the “cup” and will pervade the whole system, having only two possible outlets: through the “valve, 14,” when it is displaced, and through the nozzle, “23.” The pressure upon the diaphragm, “16” when it becomes strong enough to overcome the resistance of a “spring, 15,” will unseat the “valve, 14,” and the fluid will pass into the cylinder through the “pipe, 11” and shift the control valve, “3.” If the pressure in the steam pipe were to become strong enough to contract the bellows, “34,” the “lever, 31,” would rise and would open the nozzle, “23,” and the system would return to its original position.

If the machine had done nothing more, the result would have been a series of corrective movements of the valve, “3,” which would overcompensate: a phenomenon known in the art as “hunting.” In order to prevent this and to damp the control proportionally to changes in the steam pressure in the main pipe, Lindsay added another set of elements. Leading out of the “pipe, 11,” just as it feeds into the cylinder is a “pipe, 46,” controlled by a needle valve, 47.” This pipe ends in an “air-vessel,” “48,” from near the bottom of which the “pipe, 49,” leads to a chamber, containing a second bellows, “43,” whose collapse when the chamber is under pressure raises the “lever, 40,” against the “spring, 41.” The free end of this lever is flexibly connected with the free end of the “lever, 37”; i.e., with the “cup, 31” (page 5, lines 125-128). If the “cup, 31,” be assumed to be sealing the nozzle, “23,” and if the “valve, 14,” be unseated, the pressure in the cylinder will be communicated to the bellows, “43,” which will tend to lift the “lever, 40,” against the “spring, 41,” and push up the “cup, 31,” against the spring, “38.” The effect of this is to oppose the action of the bellows, “34”; i.e., to “damp” its action proportionally to the change of pressure in the steam pipe. This is true because in proportion as the bellows, “34,” tends to break the seal of the “cup, 31,” it also tends to reduce the pressure in the whole system and thus to reseat the “valve, 14,” in response to the “spring, 15.” That wastes the cylinder and with it reduces the pressure upon the bellows, “43,” allowing the “spring, 41,” to operate jointly with the “spring, 38,” against the pressure on the bellows, “34.” Thus every pressure upon that bellows will tend to be checked, not only by the “spring, 38,” as it would always be; but by a proportional increment of the stored force of the “spring, 41,” released by the decrease in pressure upon the bellows, “43.” It is not necessary to describe the reverse movement.

The plaintiff objects to Lindsay as an anticipation, and it is not indeed a complete one. In the first place, it operates by a “non-compressible” fluid; in the second, only the flapper is operated upon, both to correct the departure from normal in the steam pipe and proportionally to damp that correction; in the third place, the flapper is not left free to vary in its pressure upon the nozzle, but is positively pressed upon the nozzle by the joint action of the springs, “38,” and “41”; in the fourth place, the pressure upon the bellows, “43,” must pass through the “needle valve, 47.” Except for these four differences, the relevance of which to support Mason’s invention we reserve for the time being, claims two and ten of the patent would be anticipated.

In February, 1926, Wingfield filed “complete specifications,” which resulted, on July 8, 1926, in a British patent (1926, Brit. Pat. 254,469), for the purpose of obviating “hunting,” which he described as “the tendency for the controlled device to carry out its full movement” (page one, lines 71-73); i.e., to overcompensate. Wingfield proposed

Page 230

to avoid this by imparting “to the relay valve a check movement, or movement in the opposite direction to that imparted by temperature impulse device * * * so that the movement of the controlled device * * * is inhibited before it has moved through its whole travel” (page 2, lines 11-18). Figure three with its description (page 3, lines 99-130, and page 4, lines 1-23), is the nearest to Mason, and is the only part that we need discuss. Changes in temperature in the main system which, as an example, we may take as a steam chamber, are transmitted by the “thermostat, T,” set in that chamber, to the diaphragm chamber, “C,” and expand or contract it, as the temperature increases or decreases. This change in the chamber tends to raise or lower the flapper, “Q,” which seals the nozzle, “F.” As the seal is relatively tight or loose, the pressure in the “chamber, D,” is raised or lowered; that pressure being fed into the chamber through the “needle valve, E,” from the main supply. An unlettered member carries the nozzle and is fastened by means of the “extension, G,” and the “link, I,” to the “lever, K”; and any rise of the bellows in the chamber, “D,” is in this way communicated to the right end of the “lever, K,” which pivots either on the “main valve shaft, L,” or on the “spring, S.” In either case the seal of the “leakport, B,” becomes firmer. That seal is the only opening out of the chamber, “R,” and into it the controlling pressure passes unobstructed. The pressure in the chamber, “R,” thus increases, and the bellows in that chamber begins to close the main valve, “V,” reducing the temperature in the boiler.

If this were all, the “valve, V,” would completely close until the diaphragm, “C,” had collapsed, had tightened the seal of the flapper, “Q,” and the nozzle, “F,” and until the pressure in the chamber, “D,” had opened the “leakport, B.” However, the action of the bellows in the chamber, “R,” is checked by the changes in pressure within the chamber, “D,” in the following way. As the pressure in the chamber decreases, owing to the weakening of the seal between “Q” and “F,” the right end of the “lever, K,” rises and stops any “bleeding” at the “leakport, B.” This movement in its turn — with some lag depending upon the set of the “needle valve, E,” — increases the pressure in the chamber, “D,” because the nozzle, “F,” rises with the “lever, K.” This increase in pressure within the chamber, “D,” tends to depress the right end of the “lever, K,” and thus to weaken the seal of the “disc M1,” upon the “leakport, B,” which reduces the pressure in the chamber, “R,” and allows the spring upon the shaft of the valve, “V,” to withdraw the valve. Thus, although the partial closing of the valve, “V,” automatically corrects the assumed overheating in the boiler, that correction is itself checked by the bellows in the chamber, “D.” It is to be noticed that in Wingfield the damping device is not applied to the flapper, “Q,” as in Lindsay; but the position of the flapper is always fixed by the diaphragm, “C,” quite as in Mason’s Bourdon tube. Moreover, the flapper does not act against any positive pressure, like the springs, “38,” and “41,” of Lindsay; and the “nozzle, F,” is itself moved to and from the flapper by the bellows in the chamber, “D,” just as Mason’s nozzle is moved by the bellows, “39.” Of the four differences between Lindsay and Mason therefore only two exist between Wingfield and Mason: (1.) the pressure by hypothesis is that of a “non-compressible” fluid; and (2.) the pressure in the chamber, “D,” must pass through the “needle valve, E.”

On May 22, 1929, a German named Wünsch filed an application which became Pat. No. 1,959,889, issued May 22, 1934. The application was amended by the introduction of Figures 3, 4, 5 and 6, on August 16, 1930, and the plaintiff argues that these figures — and, we assume, the descriptive text — cannot be taken as earlier than Mason’s invention, because the plaintiff’s “Stabilog” was given a “field test” in January, 1930. This effort to carry back the invention cannot prevail. It rests wholly upon the testimony of Hall, unconfirmed by contemporary documents, for Exhibit P-21-N. does not support it; rather it tends to indicate that the “Stabilog” — if the document does refer to the “Stabilog” — had not yet been fully perfected; for it was only “coming along very shortly” as late as February 5, 1930. Indeed, the use of the phrase, “field test,” itself strongly suggests

Page 231

that the invention was still in fieri. In fact, there is so little difference between Wünsch’s figure one and figure three that it would not change the result if we confined ourselves to the first: but, for whatever the distinction may be worth, Figure three with its text must be taken as part of the prior art. Wünsch wished to regulate pressures in the main pipe, “D,” and he does so by means of the “valve, 21,” actuated in one direction by the pressure within the “cylinder, 19,” and in the other direction by the “spring 17¹.” Pressure changes in the main pipe, “D,” are communicated through the “pipe, 1,” to the “casing, 2,” which contains the “diaphragm, 2¹,” upon the other side of which the “rod, 14,” abuts. This rod passes through the wall of the casing and its other end abuts upon one side of the “tube, 4,” which is hinged at “4a.” When the pressure in “D” increases, the diaphragm moves the rod to the left, and the rod moves the tube to the left.

The “tube, 4,” is filled with “compressed air or a gas or an oil under pressure” (page one, lines 55, 56), coming from a source not disclosed. As the tube swings to the left under the impulse of the diaphragm, its open end moves across the orifice at the end of the “pipe, 5,” and by this means it injects into that pipe more or less of the fluid passing out of the tube. “If the pressure in the pipe, D, increases, the diaphragm 2¹ is bent to the left and swings the hollow arm, 4, to the left also so that more pressure oil will enter into the tube, 5” (page 2, lines 119-121). The “pressure oil” (which might equally well be “air” or “a gas”) enters the “pipe, 5, ” and passes to the right side of the “cylinder, 6,” in which moves the “piston, 9,” on each side of which are the springs, “7,” and “8,” so designed that the “piston is normally held in its medium position” between them (page one, lines 85, 86).

As the “piston, 9,” moves to the left it carries with it the “piston rod, 26,” which moves one end of the “double-armed lever, 11,” fulcrumed at “27,” not described. The lower arm of the “lever, 11,” pushes the “spring, 3,” in a direction opposite to that of the movement of the “piston rod, 26,” and, since the opposite end of this spring abuts upon the “tube, 4,” the motion of the “cylinder, 9,” is transmitted to the “tube, 4,” in opposition to the pressure within the “diaphragm, 2.” In this way the movement of the “piston, 9,” is “damped” and the correction of the pressure in the pipe, “D,” by means of its “valve, 21,” is proportionally held back and prevented from over correcting. Although the mechanical train disclosed is very different from Mason’s in detail, not only is the whole theory of proportional damping disclosed, but the means are in the essence the same. It is true that the pressure in the system is not controlled by a flapper obstructing the discharge of a nozzle, but the shifting of the “tube, 4,” across the mouth of the “pipe, 5,” which varies the pressure received by that pipe; and it is also true that the “pipe, 5,” does not have a lead to the “pipe, 18,” putting the “cylinder, 6,” in shunt as it were. But both these features were in Lindsay and Wingfield, and, as we shall show, a combination of elements in this narrow art was not an adequate basis for invention.

There remains the validity of Mason’s “reset” claims; supported by his disclosure: i.e., “restricted connection, 55,” and the bellows, “53,” which serve to check the “damping” action of the bellows, “39.” These two bellows are so “biassed,” that, when each is under equal pressure, they set the nozzle at a point in its range, at which, when the flapper is also set there, the temperature in the “tank, 9,” is that desired. This corrects the following difficulty. With proportioning control alone the system comes to an equilibrium when the supply of steam or other power through the “pipe, 7,” just answers the demands. Changes in the demands upon the “tank, 9,” may, however, result in an equilibrium at a point other than that of the desired temperature. It is to prevent this, and to insure that the point of equilibrium shall always be at the desired point that Mason introduced his “reset” device, which operates as follows: The fluid coming through the “pipe, 43,” is retarded in its passage through the “restricted connection, 55,” so that it reaches the bellows, “53,” later than it reaches the bellows, “39.” Eventually the pressure in the two bellows becomes equal, and the nozzle is then at the desired point. Since the flapper and the nozzle

Page 232

always tend to set the “valve, 5,” at a position which will supply the steam necessary to answer the demands on the system, the point of equilibrium will in the end be identical with the desired temperature.

Wünsch disclosed “the two chambers in the cylinder, 6,” as “connected with one another by a pipe, 6¹, into which a delicate valve, 10, preferably a needle valve, is inserted,” (page one, lines 87-89). Again, “the valve, 10, is so adjusted that a certain part-equalization of the pressures in the two cylinder chambers takes place through the pipe, 6¹, and, as the pressure in the left hand chamber of the cylinder correspondingly increases, the same takes place in the left hand chamber of the cylinder, 19” (page 2, lines 17-23). As the pressures in both chambers tend to equalize, the pressure on the “piston, 20,” increases and this tends to close the “valve, 21,” and cut down the supply. However, by this equalization the “piston, 9,” itself tends to move to the right and in so doing to move the “tube, 4,” to the left, just as the original movement of the “diaphragm, 2¹”, had done, This movement of the “tube, 4,” in turn increases the pressure in the right hand chamber of the “cylinder, 6.” Thus, we have three separate forces set up through the increase of pressure in the right side of the “casing, 2.” The first is the movement of the “piston, 9,” to the left which is communicated to the “valve, 21,” through the “piston, 20.” Simultaneously with this is the counteracting force set up through the elements 26, 11 and 3, which by moving the “tube, 4,” reduces the pressure in the “pipe, 5,” and “damps” the first effect. The third force operates later in time; and by the bleed of the “needle valve, 10,” tends to move the “piston, 9,” back to the right and in so doing to increase the pressure in the right hand chamber of the “cylinder, 6,” which as before is transmitted to the “piston, 20.” This continues until the pressure in the “casing, 2,” has been brought back to normal. Such is the undisputed testimony of Clarridge, who speaks of this feature of the disclosure as an “automatic reset.” The “needle valve, 10,” is the equivalent of Mason’s “restricted connection, 55”; the left hand chamber of the “cylinder, 6,” is the equivalent of Mason’s bellows, “53,” which tends gradually to remove the “damping” action of the “bellows, 39,” just as the left hand chamber of Wünsch gradually removes the “damping” of the train of members 26, 11 and 3.

That there are differences between the disclosures of these three prior patents and Mason’s is true, but these are irrelevant except in so far as they appear in Mason’s claims, and they do not appear there, except the element of an “elastic fluid” in all but the first three; and Wünsch disclosed “air” or a “gas” as an alternative to “oil.” We should have no warrant for limiting the claims by the elements of the disclosure which they do not include, even if the elements were new. A patentee who claims broadly must prove broadly; he may not claim broadly, and recede as he later finds that the art unknown to him has limited his invention. That is the chance he must take in making broad claims; if he has claimed more than he was entitled to, the statute does give him a locus poenitentiae, but he must seasonably disclaim the broad claims in toto. He may not keep them by interpretative limitation; he must procure new claims by reissue. This is the significance of Milcor Steel Co. v. George A. Fuller Co., 316 U.S. 143, 62 S.Ct. 969, 86 L.Ed. 1332. This applies with exceptional force to the case at bar, in which Mason not only had his day in the Patent Office, but twice thereafter amended his claims; first, by his disclaimer, next, by reissue. It is true that Mason’s single figure discloses no “needle valve” in the pipe, “43,” and the flapper, “35,” as free from any opposing force, yet from start to finish neither the claims nor the specifications so much as intimated that either of these features was an element of the invention. It would be enough to prevent such an interpretation of the claims that, after so many efforts the claims remained silent upon these details, even if the specifications had mentioned them; but the interpretation becomes doubly impossible when we remember that no invention can be saved by features which appear only in the figures, and are not mentioned in the text. Permutit Co. v. Graver Corp., 284 U.S. 52, 60, 52 S.Ct. 53, 76 L.Ed. 163; Vickery v. Barnhart, 118 F.2d 578, 585, 28 C.C.P.A., Patents,

Page 233

979. There is still another reason for the same result. We should have no warrant for supposing that, if Mason had limited his claims as the plaintiff wishes and must do, the examiner would have thought them patentable variants over Wingfield and Wünsch, neither of which he discovered. As to Wünsch, the plaintiff argues that, since it was pending while Mason’s first application was in the Patent Office, we must suppose that the examiner did not think it relevant, else he would have brought it into interference. Yet upon Mason’s original application the same examiner who later passed on the reissue, cited only two inconsequential references, although Lindsay’s American patent had already issued. It would surely be hazardous to assume that the same man who later cited Lindsay against the invention should earlier have disregarded Wunsch, if he had had it in mind.

For these reasons, merely as a matter of interpretation, we could not properly save the claims by incorporating into them those elements upon which the plaintiff relies to distinguish from the prior art; and even if we could, it would not save them. Apparently it is true that the art began to feel the need of “multiple capacity” controllers as early as 1920, and if Mason had been the first to devise one, it would be plausible to argue that so long an interval was an indication that the invention required ingenuity out of the common. But that is not the history of the development of these machines. As we have seen, Lindsay filed his “complete specifications” in November, 1924; they were certainly one step in that development. Wingfield filed his “complete specifications” in February, 1926, less than two years later; that was another step — the flapper was unrestrainedly free to respond to changes in pressure. Wünsch’s German application followed Wingfield in June, 1928; that was still another step: he disclosed the “reset,” and the alternative of “air” or “a gas,” as well as “oil.” Mason followed Wünsch in September, 1930, with whatever was new in his disclosure, and there is undoubtedly much that was new, if we altogether disregard his claims. But, were we even free to compose combinations out of such new elements in disregard of the claims, how could we say that, coming as they did and being what they are, they were inventions? Mason at best was the last of a number of inventors, who in quick succession made steps in a gradual development, and if any of the steps are notable, Mason’s does not appear to be among them. So far as he is concerned, his work is merely the latest refinement of a continuous progress, proceeding by trial and error. True, Mason may personally be entitled to more credit than that; he may have devised the whole machine unaided; but that is not the test; the law unsparingly charges him with an acquaintance with all the earlier developments of the art.

We do not forget that sheet anchor of patentees — success — and we will assume, not only that the plaintiff’s machines have had great success, but that the defendant’s as well are to be counted as further evidence of success. As one factor in determining the issue of invention, success may indeed be important; but it is usually troublesome to know what weight to give it, and it is extremely apt to lead one wholly astray. As courts have again and again observed, all sorts of factors other than the novelty and value of the disclosure may account for success: e.g., excellence of design not shown in the disclosure, fineness of workmanship, reputation of the maker, advertising and the like. In the case at bar let us, nevertheless, eliminate all such possible factors; let us suppose that it was only the differences between the disclosure of Mason and those of Lindsay, Wingfield and Wünsch, which for the first time made possible a machine acceptable to the art. We should still be as far as ever from any safe use of this evidence, for the question is of the success to be attributed to that part of the disclosure which the claims have selected. Only in case we have some assurance that the success was here due to those factors by which the claims differed from the prior art, ought we take it as a criterion; and it would certainly be a gratuitous assumption to suppose that the industry’s acceptance of these machines was due in any measure to those factors in the claims, if indeed there are any. We therefore reject the industry’s acceptance

Page 234

of these machines as any evidence whatever that Mason’s claims were valid; and, being thrown back upon any advance over the art discoverable in them, we are at a loss to find any unless it be the gathering of factors which had all appeared before. When that occurs by gathering from remote arts, it is often an index of much insight and perspicacity; but not when the combination is of factors to be found in the very field, and especially in so special a field as this; and certainly not when the tempo has been so swift as it was here.

Finally, when all else is said, we cannot ignore the change in recent years in the standard of invention adopted by the Supreme Court.[4] It is as idle to pretend that there has been no change (In re Shortell, 142 F.2d 292, 31 C.C.P.A., Patents, 1062), as it would be to protest against it; and it is as much the duty of a lower court to give effect to it, as it is its duty to give effect to any other of the decisions of that court. We are to dispose of the question of invention, so far as we can divine, as we think the Supreme Court would dispose of it; and, once the cloud of complicated scientific theory, with which it has been surrounded, is blown away, and the issues are bared as they really are, we have no doubt that that court would hold the claims invalid.

Judgment reversed; complaint dismissed.

[1] 307 U.S. 243, 59 S.Ct. 857, 83 L.Ed. 1264.
[2] Permutit Co. v. Wadham, 6 Cir., 13 F.2d 454, 456; N.O. Nelson Mfg. Co. v. F.E. Myers Bro. Co., 6 Cir., 56 F.2d 512; United Chromium, Inc., v. International Silver Co., 2 Cir., 60 F.2d 913; Lowell v. Triplett, 4 Cir., 77 F.2d 556, 561 (semble); Payne Furnace Supply Co., Inc., v. Williams-Wallace Co., 9 Cir., 117 F.2d 823, 827; Bay State Optical Co. v. Klein, D.C.E.D.N.Y., 20 F.2d 915, 917, affirmed 2 Cir., 23 F.2d 999.
[3] 307 U.S. 243, 59 S.Ct. 857, 80 L.Ed. 1264.
[4] Buono v. Yankee Maid Dress Corp., 2 Cir., 77 F.2d 274, 276; Picard v. United Aircraft Corp., 2 Cir., 128 F.2d 632, 636; Aero Spark Plug Co., Inc., v. B.G. Corp., 2 Cir., 130 F.2d 290, 292; United States Gypsum Co. v. Consolidated Expanded Metal Companies, 6 Cir., 130 F.2d 888, 892.

On Petition for Rehearing.
PER CURIAM.

The petition for rehearing points out a mistake in our description of Wingfield’s disclosure: The “spring S” is a tension spring and cannot act as a pivot for the “lever K.” Next it is true that when the left end of the “lever K” descends with any increase of pressure in the chamber “R” the “leakport B” will tend to open and this is what decreases the pressure in that chamber and stops the descent of the valve. On the other hand it is equally true that as the “diaphragm C” by expanding raises the flapper, “Q” and thus decreases the pressure in the chamber “D” the right end of the “lever K” rises, thus tending to tighten the “leakport B.” There is in this way a continual balancing between the forces acting upon the “leakport B” which is the important thing.

We find nothing in the petition to change our original disposition of the appeal, and it is denied.

Petition for rehearing denied.

Page 235