Full opinion text
CHESNUT, District Judge. The three patents involved in this equity suit all relate to the annealing of glassware, and more particularly hollow glassware, such as bottles. Annealing is a process of heating and cooling. The pleadings in fact present for consideration six of the plaintiff’s patents but counsel have agreed to rest the case upon three and to limit the number of claims in each to those hereinafter enumerated. The three patents and the particular claims thereunder for consideration are as follows: Mulholland reissue, No. 17,263, April 9, 1929, claims Nos. 2, 7, 44, 49, 50, and 52 to be taken as typical; Ingle No. 1,583,-046, May 4, 1926, claim 5; Mulholland No. 1,840,463, January 12, 1932, claims 1 to 5, inclusive, with claim 5 typical. The plaintiff is a Connecticut corporation which designs, builds and licenses the use of lehrs for the annealing of glassware. The defendant, Swindell Bros., Inc., is a Maryland corporation with plant in Baltimore City, manufacturing glass bottles. It uses three of the plaintiff’s lehrs under license agreement and it also owns and uses two lehrs designed and made by the Amsler-Morton Company, a Pennsylvania corporation, which has been permitted to intervene in the case and as maker of the alleged infringing machine, has assumed the burden of the defense of this patent infringement suit for injunction and accounting. The emphasis of the defense is placed on the point that the intervenor’s apparatus does not infringe that of the plaintiff, but if infringement should be found, then the invalidity of the plaintiff’s patents in view of the prior art is advanced as a further defense. As to the latter, the plaintiff replies that the defendant (and consequently the intervenor also) is as licensee of the plaintiff’s apparatus, and by virtue of a special stipulation in the license agreement, estopped to deny the validity of' the plaintiff’s, patents; but the intervenor says this position is not tenable because the alleged infringing device is entirely outside the scope of the patent agreement and therefore the estoppel does not arise. It may be said at once 'that it will not be necessary to adjudicate this issue in this case because I have reached the conclusion that there is no infringement, ' In passing it may, however, be pointed out there is substantial support for the intervenor’s position in the very persuasive reasoning of Judge Dennison in the case of Indiana Mfg. Co. v. Nichols & Shepard Co. (C.C.) 190 F. 579, which was adopted and applied by Judge Sanborn in the later case of Symington Co. v. National Malleable Castings Co. (D.C.) 257 F. 564. See, also, International Burr Corp. v. Wood Grinding Service (C.C.A.2) 34 F.(2d) 905; Eskimo Pie Corp. v. National Ice Cream Co. (C.C.A.) 26 F.(2d) 901; Paramount, etc., Co. v. Moor-head, etc., Co. (D.C.) 251 F. 897, 902, affirmed (C.C.A.) 260 F. 841; Good Humor Corp. v. Bluebird Ice Cream (D.C.) 1 F.Supp. 850, affirmed (C.C.A.) 66 F.(2d) 1013; and Tate v. B. & O. R. Co. (C.C.A.4) 229 F. 141. And it is not denied by plaintiff’s counsel that Swindell as the defendant in this case may refer to the prior art for the construction of the scope of the plaintiff’s patents in support of defendant’s claim of noninfringement. Westinghouse E. & Mfg. Co. v. Formica Insulation Co., 266 U.S. 342, 45 S.Ct. 117, 69 L.Ed. 316. The intervenor for itself and for the defendant raises the defense of laches on the part of the plaintiff in the institution of the suit. It is pointed out that the Amsler-Morton Company’s apparatus has been extensively in use in its present form since early in 1927; that the plaintiff became substantially advised as to its structure and operation at least’ as early as 1928 and thereafter on July 26, 1928, filed its application for its reissue patent No. 17,263, for the purpose of more clearly, in one or more of the claims, covering the Amsler-Morton lehr without successful defense based on a prior use’ by the Amsler-Morton Company in a lehr built for Frey & Son in 1919; and that the reissue patent as applied for was issued on April 9, 1929, and shortly thereafter the plaintiff charged the defendant with infringement under the reissue patent, which was denied by the Amsler-Morton Company’s patent counsel in an extended letter giving the reasons for the denial. It is said the issue between the plaintiff and the Amsler-Morton Company was thus definitely joined on the question of patent infringement but nevertheless no suit has yet been brought by the plaintiff directly against the Amsler-Morton Company up to the present time. The plaintiff, however, takes the position that the defense of laches is a personal one to th'e defendant, Swindell, and therefore any delay in suit by the plaintiff against the Amsler-Morton Company is not properly in issue in the case. See Chandler & Price Co. v. Brandtjen & Kluge, Inc., 296 U.S. 53, 56 S.Ct. 6, 80 L.Ed. 39. And as to Swindell, it is pointed out by the plaintiff that the former acquired the alleged infringing apparatus only in April, 1931, and .that suit was instituted here on July 30, 1934. Plaintiff’s counsel offers an explanation and excuse for such delay as occurred by reason of lack of full and precise information as to the Amsler-Morton apparatus until 1930, and lack of reasonable opportunity to sooner sue due to other patent litigation. The theory of the intervenor here is that the plaintiff’s long delay and failure even up to the present time in suing the intervenor operates as an implied license to it which it can validly pass on to Swindell. It is certainly doubtful whether laches alpne without some positive element of estoppel, such as. change in position, would be so operative. The plaintiff’s excuses for delay in suing the Amsler-Morton Company are not impressive, but it would seem that even if laches were sufficient in this case to preclude an accounting as against Swindell, it would not be sufficient to deprive the plaintiff of its right to an injunction if otherwise existent. See Menendez v. Holt, 128 U.S. 514, 524, 9 S.Ct. 143, 32 L.Ed. 526; Denominational Env. Co. v. Duplex Env. Co. (C.C.A.4) 80 F.(2d) 186, 193; A. R. Mosler & Co. v. Lurie (C.C.A.2) 209 F. 364. It is, however, unnecessary to determine this issue of laches in this case in view of the conclusion already announced with regard to infringement. Before detailed consideration of the plaintiff’s several patents, and comparison of its apparatus with the alleged infringing apparatus of the Amslcr-Morton Company (hereinafter some times for brevity called “Amco”), it is desirable to state very briefly, but sufficiently for the purpose of understanding the patents in the case, the essentials of the manufacture of glassware. The art is indeed a very old one. It was practiced to some extent by the Egyptians, more largely by the Romans, quite extensively in the Middle Ages in Western Europe, and also in this country during the Colonial period, and at all times subsequent thereto. Very briefly, the raw materials for glass making are melted in a large furnace or retort at a very high heat (about 2,500 degrees Fahrenheit) and the portions of the viscous mass are poured into molds or form-shaping machines at a temperature of about 1,800 degrees. From these molds the glassware thus formed, still in a plastic state, is transferred either manually or by an automatic conveyor a short distance into a tunnel-like structure called a lehr, about 60 to 70 feet in length, which is heated at the receiving end to a temperature of approximately 1,000 degrees, with a temperature gradient in the end. The glassware is transported slowly through the lehr on an automatic conveyor belt. The function of the lehr is to first reheat the glass and then gradually and uniformly cool it. This is called annealing. Plaintiff’s patents all relate to the structure of the lehr made by it with one or more claims also for the method of annealing involved. When the hollow glassware, in bottle shape for instance, is removed from the shaping molds the outside, which is in contact with the mold, is somewhat cooler than the inside and therefore the relatively greater contraction of the glass on the outside sets up strains and stresses which, if the glassware were allowed immediately thereafter to cool at the normal factory temperature, would cause it to at once break as a result of its own internal forces or to readily break thereafter upon slight pressure. To obviate this it is necessary that the glassware should be quickly placed in the lehr where for a certain period of time it will be again subjected to a relatively high temperature sufficient to renew its plasticity, and then it is allowed to cool in the slowly decreasing temperature until the whole of the glass article has become permanently set by uniform and gradual cooling. The range of temperature which must be maintained during this annealing stage in the manufacture of glassware is from 1,000 degrees F. to approximately 750 degrees. It is essential, therefore, that this temperature be maintained in the lehr during the annealing stage. It is called the critical range of temperature. The linear extent of the lehr in which this critical range of temperature must be maintained depends upon the amount and thickness of the glassware and is, of course, also affected by the rate of forward progress of the glassware through the lehr. After the glass has successfully passed the annealing stage the remaining function of the lehr is merely to gradually reduce the temperature during the further progress* of the glassware until if is at a sufficiently low temperature to be handled at the delivery end of the lehr; and after inspection, packed for shipment. The particular range of temperature at the receiving end of the lehr is therefore not so vitally important as that to be maintained during the annealing stage; but is of course important for the gradual reduction of temperature. In commercial practice, for efficiency and economy of operation, it is obvious that the most desirable characteristics for a lehr are economy in fuel, which however must be sufficient to maintain the required temperature during the annealing stage, and rapidity of progress through the lehr, in order to speed up mass production. The history of the art of glass making, over the last 50 years, shows a steady improvement in construction for economy and efficiency of operation of lehrs. The earliest form of apparatus for annealing was merely a closed oven in which the glass was placed after formation in desired shape by blowing, and while it was practically cold. The oven was then heated to the required temperature which was maintained as long as necessary, after which the heating was discontinued and the glass allowed to gradually cool before it was manually removed from the oven; which had been sealed during the operation. This process was entirely efficient as to the quality of the annealing, but was necessarily slow in operation. With the introduction of machinery for shaping the glass, there came also the necessity of speeding up the annealing process for mass production. The introduction of the lehr was the response of the trade to this demand. The earliest form of lehr in use about 50 years ago, consisted of a brick structure forming the exterior walls, necessarily placed for stability on firm foundations, an open gas flame within the forepart or receiving end of the lehr and some form of conveyor operated either by hand or machinery for the transportation of the glassware through the tunnel. As compared with present practice, this early form of lehr had numerous disadvantages in that the products of combustion of the open flame often discolored the glassware, the rate of progress through the lehr was very slow, varying from 4 to 8 hours, and the absence of an efficient control of the temperature gradient resulted in a comparatively high percentage of breakage of the glass during annealing, which varied from 5 to 10 per cent. These lehrs were known as “open fire” lehrs. With marked improvement in the molds or forming devices and the speeding up of mass production came also improvements in the design, structure and efficiency of the lehr. The first important change was to enclose the products of combustion from the fire in flues which were variously arranged around the sides and length of the lehr. This prevented the discoloration of the glassware, and by the use pf dampers variously spaced in different part's of the flues afforded a means for some control over the temperature in the lehr. This improvement came into general use shortly after the turn of the century. The next great advance came during or shortly after the World War and is illustrated by the types of lehrs constructed by the Amsler-Morton Company, in the installation made by it for H. C. Frey & Sons, manufacturers of glassware in 1919, and by the Dixon Glass Company for the Capstan Glass Company, and for the Owens Bottle Company of Charleston, W. Va. These lehrs, had outlets for the flues (which were both longitudinal and vertical) at various intervals in the length of the lehr, which in turn were controlled by dampers, so that the distribution of the heat from the firebox in the forepart of the lehr could be more efficiently regulated throughout that portion of the lehr in which the annealing of the glassware occurred. In the meantime improvements had been made in the form of the conveyor belt used. These new lehrs added greatly to the efficiency of production, whereby the amount of breakage and necessary rejection was reduced to about 2%, and the time required for annealing and discharge from the lehr was very materially reduced. This type of lehr, however, was still of the fixed stationary variety with external walls of brick. The next marked improvement was the introduction about 1920 of the movable type of lehr which is now used both by the plaintiff and by the Amsler-Morton Company. This type of lehr has instead of brick walls sheet metal heavily insulated with rock wool. It is much lighter in construction than the brick lehr and generally is narrower in width (being abo.ut 3 to 4 feet wide while the brick lehr was about 10 feet wide), is equipped with a more rapidly operating conveyor belt, and by reason of its smaller size and less weight, is capable of location at will in relation to the forming machines. In the trade it is generally called the “unit” type of lehr. It is generally about 70 feet in length. First: In construing the "plaintiff’s patents and determining the question of infringement, it is to be importantly noted that we are not dealing with pioneer patents but with those that represent only improvements on an apparatus in an art that is very old and in a field that is literally crowded with prior patents. The doctrine of patent law which is applicable to this situation is well and succinctly summarized in section 230 of the 6th Ed. of Walker on Patents, Vol. I, where the author states the principles to be derived from the four leading cases of McCormick v. Talcott (1857) 20 How. 402, 15 L.Ed. 930; Chicago & N. W. Railway Co. v. Sayles (1878) 97 U.S. 554, 24 L.Ed. 1053; Morley Sewing Machine Co. v. Lancaster (1889) 129 U.S. 263, 9 S.Ct. 299, 32 L.Ed. 715, and Kokomo Fence Machine Co. v. Kitselman (1903) 189 U.S. 8, 23 S.Ct. 521, 47 L.Ed. 689: “The meaning of these four cases seems to be that every inventor is entitled to claim whatever he was the first to invent. If A. B. is the first to invent mechanism to perform a particular work, and if his mechanism is substantially incorporated into subsequent machines which do that work, then A. B. is entitled to such a construction of his patent as will be infringed by those later machines; but if C. D. is a mere improver on A. B.’s machine, C. D. is not entitled to such a construction of his patent, as will cover the machines of still later inventors, who have improved on A. B.’s machine in a substantially different manner. It follows from these doctrines that C. D.’s patent must be construed in the light of A. B.’s machine, and indeed in every other similar and older structure; which is the same thing as saying that every patent must be construed in the light of the state of the art, at the time the invention it covers was produced.” And again: “The doctrine of the Sayles Case is as follows: If one inventor, in a particular art, precedes all the rest, and strikes out something which underlies all that they produce, he subjects them to tribute. But if the advance toward the thing desired is gradual, so that no one can claim the complete whole, then such inventor is entitled to the specific form of device which he produces, and every other inventor is entitled to his own specific form, so long as it differs from those of his competitors and does not include theirs.” And still again: “But if the invention claimed, be itself but an improvement on a known machine, by a mere change of form or combination of parts, the patentee cannot treat another as an infringer, who has improved the original machine, by use of a different form or combination, performing the same functions. The inventor of the first improvement cannot invoke the doctrine of .equivalents to suppress any other improvement which is not a mere colorable invasion of the first.” With this general background of the art we come to a particular consideration of the first of the three patents in suit. It is the MulhoUand reissue patent No: 17,263. We are told by the patentee in the opening of his specifications that his invention relates to a method of and an apparatus for annealing glassware and more specifically to the annealing of glassware in a lehr of the tunnel type in which the ware passes in procession through the tunnel whose temperature varies from end to end. Something of the history of the art is then set out and the patentee proceeds to state the objects of his invention to be in general: “To produce better annealing of glassware, to save as much fuel as possible in the annealing, and to reduce the time of annealing as well as to overcome or minimize the disadvantage of tunnel lehrs. More specifically, the object is to create a proper environment through which the glassware may be moved continuously on a conveyor and in which it may, be subjected, as nearly as possible, to the ideal cooling conditions to produce the best possible annealing and to reduce the time and cost of operation.” Other objects are stated to be improved means and methods for controlling the temperature throughout the length of the lehr, to overcome uncontrolled air currents within the lehr; and to control the local convection currents within the tunnel and utilize them to produce the desired thermal environment and as nearly as possible a uniform temperature for the glassware. To reach his objective he states that he provides a lehr “having a plurality of lower flues extending longitudinally beneath the tunnel and a plurality of upper flues extending longitudinally above the tunnel, all of said flues being provided with damper-controlled . openings at spaced intervals throughout their length and communicating with the outside atmosphere.” It is further pointed out that in operation the heated gases from the firebox at the receiving end of the lehr pass through the lower flues and by operation of the dampers at spaced intervals the amount of heat delivered in various areas or zones of the longitudinal length of the flues can be measurably controlled by opening the dampers in the flues to let in cool air from the factory atmosphere, so that of course the temperature in the flues containing the hot air can be lowered. And a further reduction of temperature is accomplished by the cooler factory air being sucked or blown into the upper flues and in turn measurably regulated by their damper control openings at spaced intervals whereby mire or less of the cool air is allowed to escape to the outside atmosphere at selected intervals. It is to be noted also that the heated air travels from the receiving end of the lehr toward the delivery end while the cool air travels in the opposite direction. The range of temperature in the heated flues is also subject to regulation by the volume and intensity of heat generated in the firebox and provision is made there also for the introduction of a desired amount of outside air to be mixed with the products of combustion. The amount of temperature in the lehr is therefore affected by two factors: (1) The positive amount of heat generated in the burner and passing through the heated flues and radiated of course therefrom into the tunnel of the lehr which, however, may be reduced by the amount of cooler air let in from the valve in the flue placed just to the rear of the firebox, thus diluting the products of combustion, and may be further reduced by opening the dampers in the inlets to the flue spaced much further to the rear of the flue and beginning in practice near the end of the annealing stage within the lehr; and (2) by the wholly cool air sucked into the upper flues which in turn may be reduced in length of operative effect by the opening or closing of the spaced dampers in the outlets from the upper flue.- As .has been pointed out, what is desired in operation is a high temperature at the receiving end of the tunnel to be maintained for a certain part of the length .of the tunnel at a temperature not less than 750 degrees, and a relatively lower temperature gradually decreasing through the remaining length of the tunnel; and some approximate control over the length of the zone in which this critical range of temperature for the annealing of glassware is required, to be regulated through the operation of the dampers whereby the length of the flue containing the highest heat measuring from the firebox can be lengthened or shortened, and similarly the check on the higher temperature and consequent lowering of temperature measuring from the delivery end may be influenced by the amount of cooling air in the upper flues in which the air current moves in the opposite direction. Thus the higher heated zone within the tunnel from the receiving end may be lengthened or shortened as desired for the particular glassware that is being annealed. This relation of upper and lower flues carrying respectively hot and cold air, it is claimed, has the incidental advantage of causing the heated air to rise from the bottom to the top of the lehr, thus establishing convection currents in a vertical direction only and tending to prevent the longitudinal air currents flowing in from the cool delivery end toward the heated receiving end, which was a disadvantage of the older form of lehrs because it tended to destroy the uniformity of temperature and to disturb its general decline from the receiving end to the delivery end. Further elimination of these longitudinal air currents is also aimed at by the feature of the arrangement of the conveyor belt which, when entirely within the tunnel as to both the upper and lower surfaces, likewise tends to create longitudinal 'air currents. In plaintiff’s lehr two things are done with respect to the conveyor belt to obviate this tendency; one is to incline the conveyor belt slightly from the receiving end toward the delivery end; and the other is to have only the ware-bearing surface of the conveyor belt within the tunnel, the return strand passing underneath the tunnel. The relative locations and the inter-action of the longitudinal hot and cold air flues constitutes an important feature of the structural arrangement of the lehr according to the specifications of the patent. Additional details of construction as indicated in the specifications are that the belt of the endless conveyor is composed of woven wire fabric, the ware-bearing strand of which rests upon the flat floor of the tunnel; and it is said: “By reason of the light weight and open construction of the conveyor, it has low heat capacity and relatively greater surface area in proportion to its mass, and therefore follows very closely the temperature of. its environment.” And it is also noted in the specifications that “the roof of the upper wall of the tunnel is constructed of metallic sheets provided with a plurality of parallel longitudinal extended corrugations substantially V shaped in cross-sections. These corrugations enclose the cooling flues mentioned above, for conveying cooling air adjacent to the top of the tunnel, and they also provide a roof for the tunnel having a relatively large heat-absorbing surface in comparison to the width of the tunnel, and their formation also tends to equalize the temperature transversely of the tunnel.” The drawings accompanying the application for the patent also disclose these particular features of construction. It is apparent from the file wrapper of the patent that the patentable difference over the prior art which induced the granting of the patent is attributable very largely if not entirely to that feature of the patent which provides for the dilution of the hot gases in the lower flue by the inlets of outer air at spaced intervals controlled by dampers as previously explained. And it should be also noted that practically every feature of this particular construction has been anticipated in some prior patent or prior use, as disclosed by the testimony in this case as to prior art, with the single exception of the dilution of the hot gases by the outer air inlets. With this analysis of the patent we come now to a consideration of the construction and scope of the claims in suit. Claims 1 and 2 are for the method of annealing glassware and claims 7, 44, 49, and 52 are for the apparatus. It will be sufficient to consider the apparatus claims as there is no significant difference between the method and the function of the apparatus. Plaintiff’s counsel submit the apparatus claims, Nos. 7, 44, 49, and 52 as typical or example claims. They are as follows: “7. A lehr for annealing glassware comprising a tunnel, a flue associated with said tunnel, means for causing a heating medium to flow through said flue, and means for reducing the temperature i/n said flue selectively at any one of a plurality of intervals along said flue, and for thereby changing at will the temperature within said tunnel. “44. A lehr for annealing glassware comprising a tunnel, a flue associated with said tunnel, means for causing a heating medium to flow through said flue, the said flue being divided into a plurality of independently controllable heating zones, and means, including spaced dampers, for increasing or decreasing, at will, the temperature in any of said zones. “49. A lehr for annealing glassware comprising a tunnel, a longitudinal flue extending beneath said tunnel, means for causing a heated medium to flow through said flue to heat directly the bottom only of said tunnel, said flue being divided into a plurality of independently controllable zones, and means for varying the temperature drop in contiguous zones from the hotter toward the cooler end of said tunnel. “52. A lehr for annealing glassware comprising a tunnel, a conveyor of low heat mass movable through said tunnel for transporting the ware therethrough, while permitting sensitive control of the transference of heat to and from the ware, a flue associated with said tunnel and adapted to conduct a heated medium there-through, said flue having a plurality of contiguous controllable temperature controlling zones, means for causing a heated medium to flow through said flue, and means including spaced dampers for independently controlling at will the temperature drop in any of said zones, whereby the temperature of the ware may be accurately controlled by temperature of the medium in the flue, and whereby the ware is made quickly responsive to such control.” The critical phraseology in each of said claims has been italicized for clarity. Analyzing these claims we find that No. 7 is perhaps the broadest in its literal scope as it does not define the means for reducing the temperature in the heated flue at selected intervals. Claim 44 is a little more specific in that it expresses that the means for controlling temperature include spaced dampers. Claim 49 defines the heated flue as “a longitudinal flue extending beneath said tunnel whereby the bottom only of the tunnel is heated.” Claim 52 is moré specific as to the plaintiff’s actual structure. It introduces an element of the conveyor belt as “a conveyor of low heat mass movable through said tunnel for transporting the ware there-through while permitting sensitive control of the transference of heat to and from the ware.” Otherwise it is substantially similar to claim 44. The conveyor referred to is the woven wire belt for moving the glassware and will be more particularly considered under the Ingle patent in suit. Claim 44 was claim 51 of the Mulholland patent which was originally issued as patent No. 1,571,137, January 26, 1926. It became claim 44 in the reissue patent; and claim 52 of the reissue patent is new to it. It appears from the file wrapper of the reissue patent as further explained by counsel for the plaintiff, that the occasion for applying for the reissue patent was that after its original issue on January 26, 1926, and before applying for the reissue patent, patent counsel for the plaintiff on inquiry from the Amsler-Morton Company, learned that the latter contended that its installation of a lehr at the plant of Frey & Company in 1919 was a prior use which invalidated the claim 51 of the original patent, now 44 of the reissue patent, and therefore claim 52 was drawn to be more specific by including in the patentable combination the element of the specially described belt conveyor. It will be noted that none of these claims include as a feature of the patent the important element of the cold air flue which is stressed in the specifications. Nor is any claim made for the so-called unit type constructed lehr as contrasted with the older stationary type; nor is there any claim for the structural features of the unit type of lehr consisting in the substitution of sheet metal for the older brick wallfe, nor for the insulation thereof. All these features are clearly shown by the testimony to have been anticipated in the art long prior to the patent in suit. The only features of the claims which are involved in this issue of infringement are (1) the use of the longitudinal lower flues; (2) the dilution of hot air therein by the hampered inlets from the outer air at spaced intervals; and (3) the nature and type of the conveyor belt. The last of these will be considered more fully in connection with the Ingle patent. It is sufficient here to say with regard to the particular patent now being considered that the defendant’s conveyor belt, although of woven wire, is differently constructed and applied in the lehr from the plaintiff’s arrangement. In argument plaintiff’s counsel conceded that no claim of the patent in suit is broad enough to base infringement only on defendant’s use of the lower flue which indeed is clearly shown to have been previously used in the art. Indeed it may be said that the testimony shows in prior use or disclosure almost every conceivable variety of arrangement of flues in these lehrs, both lower and upper, horizontal and vertical, communicating and separate. And the testimony as to the prior art shows very clearly that practically every feature of this reissue patent has been anticipated with the single exception possibly of the cool air inlets for the heated flue. The prior art in this case is very fully shown in the extended testimony of the defendant’s expert witness, Schwartz. After fully describing the patent in suit, and comparing the claims in issue with the defendant’s apparatus, he very briefly and succinctly summarized his testimony by saying: “I will just say briefly, in summing up, that the principles found in the defendant’s lehr of passing hot gases longitudinally through a tunnel which has an endless conveyor traveling over that flue, and the taking off of gases at intervals along the tunnel, is found in the defendant’s earlier lehr at the Frey Glass Company, it is found in the Milner patent and in the lehrs built in accordance with the Milner patent, at the Capstan Glass Company and at the Owens Bottle Company; it is also found in the Cruikshank patent and in the Pike patent, and it is found in the Hemingray installation of the defendant,” — all more than two years prior to the application for the patent in suit. He continued, referring to the prior uses mentioned and particularly to the outlets from heated flues at spaced intervals in the former installations, which are comparable to similar devices in the defendant’s structure, as follows: “Now by those uptakes, those methods of regulation, the temperature gradient and the annealing curve along the tunnel could be modified at will by those .dampers that were in those uptakes.” Other patents specifically relating to lehrs for annealing glassware and printed publications more than two years prior to the application for the reissue patent very fully show that it was old in the art to use a heating flue longitudinal of the length of the lehr in some cases placed at the bottom of the lehr and underneath it and in other cases at the top; so that particular feature of claim 49 was of itself'by no means new in use by Mulholland. In my opinion the claims of the reissue patent relied on and above quoted must in respect to the means of heating the lehr and controlling the temperature in the tunnel, be limited by construction to the means therefor described in the specifications of the patent and emphasized by the patentee as shown by the file wrapper as a basis for the establishing of a patentable difference from the prior art, that is, to cold air inlets for the lower hot flue, with damper control. As was said by Judge Soper in this court in Standard Brands, Inc., v. Federal Yeast Corp., 38 F.(2d) 329, 338: “Meritorious patents are frequently saved by interpreting the language of the claims so as to include restrictions on limitations appearing in the specification. Bisight Co. v. Onepiece Bifocal Lens Co. (C.C.A.) 259 F. 275, 276; Van Ness v. Layne (C.C.A.) 213 F. 804, 807; Fowler & Wolfe Mfg. Co. v. McCrum-Howell Co. (C.C.A.) 215 F. 905, 909; Permutit Co. v. Wadham (C.C.A.) 13 F.(2d) 454, 458; Gibbs v. Triumph Trap Co. (C.C.A.) 26 F.(2d) 312; Smokador Mfg. Co., Inc., v. Tubular Products Co. (C.C.A.) 31 F.(2d) 255, 257.” And in the later case of Denominational Envelope Co. v. Duplex Envelope Co., 80 F.(2d) 186, 192, it was also said by Judge Soper, speaking for the Circuit Court of Appeals, that: “Of course the claim as allowed must be read and interpreted with reference to the rejected claims, and to the prior state of the art, and cannot be construed to cover what was rejected by the Patent Office or disclosed by prior devices. Doughnut Machine Corporation v. Joe-Lowe Corp. (C.C.A.) 67 F.(2d) 135.” We come now to the question whether the Amco apparatus used by Swindell infringes plaintiff’s reissue patent. The patent statutes do not define what constitutes infringement, but many judicial decisions have established the doctrine so that there is no uncertainty in the law as applicable to this case. To constitute infringement there must be identity of function and substantial identity of way in performing that function; or in slightly different wording, a device infringes “if it performs substantially the same function in substantially the same way to obtain the same result.” Sanitary Refrigerator Co. v. Winters, 280 U.S. 30, 42, 50 S.Ct. 9, 13, 74 L.Ed. 147; Westinghouse v. Boyden Power-Brake Co., 170 U.S. 537, 18 S.Ct. 707, 42 L.Ed. 1136; Skelton v. Baldwin Tool Works (C.C.A.4) 58 F.(2d) 221; Hoeltke v. C. M. Kemp Mfg. Co. (C.C.A.4) 80 F.(2d) 912, 922; Oates v. Camp, 83 F.(2d) 111 (C.C.A.4). The flexibility of the rule depends upon the range of application given to the word “substantial.” It is of permissibly broader scope in enforcing primary patents than where the latter are merely secondary, that is for improvements of a machine, as in the instant case. While any patent is entitled to “some range of equivalents” (Oates v. Camp, supra) “the range of equivalents depends upon and varies with the degree of invention” (Paper Bag Patent Cases, 210 U.S. 405, 415, 28 S.Ct. 748, 749, 52 L.Ed. 1122). As the claims in suit clearly do not mark any high degree of invention in view of the prior art, the patent in suit is not entitled to a broad range of equivalents, that is, the word “substantial” as here used is in this case to be strictly rather than broadly applied. Walker on Patents (6th Ed.) § 416, p. 510; Frick Co. v. Lindsay (C.C.A.4) 27 F.(2d) 59, 62. Models of the competing apparatus have been offered in evidence. They are both movable lehrs of the modern unit type. Their exterior walls are both of sheet metal heavily insulated with rock wool. They are of substantially similar exterior appearance with the exception that the defendant’s lehr is somewhat larger and wider. But the exterior structure is not involved in the patent claims and is of no particular significance in the case. The question as to infringement arises on the interior structure including the arrangement of the flues and the conveyor belt. So far as the conveyor belt is concerned, that will be considered in connection with the Ingle patent. , As to the flue arrangement, we find the only real similarity between the two structures to consist in the fact that both have a longitudinal lower flue beneath the lehr. The defendant’s structure, however, also has an upper flue of short length and it has separate burners for each flue. The striking difference in function and operation of the defendant’s apparatus as compared with the plaintiff’s results from the fact that it does not have the plaintiff’s feature of the dilution of the heated gases in the lower flue at spaced intervals by damper openings permitting the introduction of cool air into the heated flue. Defendant’s apparatus is, however, provided at spaced intervals with dampered uptakes from the flues whereby the heated gases can be let out through a manifold into the outer air. There are four of such outlets in the defendant’s structure, and the temperature control in the annealing stage of the lehr is effected by the opening or closing of the dampers whereby the intensity of heat at spaced intervals from the burners can be to some extent regulated by the manipulation of the several dampers. It is to be noted that the heated lower flue of the defendant’s structure extends longitudinally only about two-thirds of the length of the lehr, the remaining one-third being without positive heat control. The latter is the portion of the tunnel which the glassware traverses after leaving the annealing stage. Defendant’s structure contains no cold air flues óf any kind' but does have in the rear third of the lehr raisable openings in the top something like hinged trap doors, which further operate to permit the escape of the heated air, and in this way, by subtraction of heat, tend to lower the temperature in the rear end zone. The striking functional difference between the means employed by the plaintiff and by the defendant respectively to measurably control temperature within the lehr is that the plaintiff’s apparatus xunctions in that respect by the introduction of cold air, while the defendant’s apparatus operates by the subtraction or escape of hot air. As might be readily understood from general knowledge with regard to heating and cooling devices for various enclosures, houses and others, and operation of stoves and furnaces, there is nothing new in the use of dampers to control the flow of heat. And it is not surprising, therefore, to find in the art of glass annealing that lehrs for many years have been provided in various ways with damper control of heated flues. We find this feature of temperature control in many of the prior patents for lehr construction and referred to in printed publications. In general it may be said that the Amco lehr uses no method of heat control or regulation, as compared with the plaintiff’s structure, and its patent claims, in suit, which was not known and old in the art more than two years prior to the plaintiff’s patent application. In addition to this essential difference between the method ot temperature control in the lehr between the two structures, there are numerous other structural differences between the two lehrs. These have been succinctly summarized by defendant’s expert witness Schwartz as follows : The attrition of the trial and argument has narrowed the issue as to infringement of the reissue patent to a single point. Do the outlets from the defendant’s lower heated flue at spaced intervals by dampers constitute a substantial equivalent for the inlets at spaced intervals of cooler air which dilutes the hot air flue of the plaintiff’s structure as provided for in the patent? I conclude there is no such infringement for at least three reasons. (1) . There is a substantial difference in function and operation between outlets of hot air from flues into the outer air and inlets of cold air into the hot gases. One is subtraction of heat, the other is dilution of heat. Both result in diminution of heat but the result is accomplished in opposite ways. This difference between the two methods was finally successfully urged by plaintiff’s counsel in the prosecution of the patent in the Patent Office as appears from the file wrapper. It is not suggested that the plaintiff is now es-topped by this successful argument in procuring the patent, but the distinction seems sound to me as an answer to infringement. (2) . The primary function of the plaintiff’s inlets is not to control the temperature within the annealing zone in the lehr, but in the cooling zone; while the defendant’s outlets are operative only or certainly primarily within the annealing zone. This reason requires some further . detail of the facts. It will be observed from the drawings of the plaintiff’s patent in suit that the inlets are spaced toward the rear portion of the tunnel of the lehr while the defendant’s outlets are placed much nearer the front of the lehr. The annealing process as a whole consists of three stages: (a) A short period where the glassware in the receiving end of the lehr is re-heated to the plastic stage at a temperature of about 1,000 degrees; (b) it passes then into what is more exactly the annealing stage of gradual cooling at a desirable temperature, the critical range of which is from 1,000 degrees to 750 degrees; and (c) the final stage is further gradual cooling to the end of the tunnel. The defendant’s flue does not extend longitudinally beyond the annealing stage, while the plaintiff’s heated flue does extend the whole length of" the tunnel. Charts of the temperature changes in the respective lehrs during actual operation at the Swindell plant show that the annealing range has been passed in the tunnel of the plaintiff’s lehr at a considerable distance prior to the point where the glassware comes opposite the first of the inlets, and an instruction book issued by the plaintiff for the operation of its lehr also indicates that the annealing stage is completed before the glassware reaches the first of the inlets. Plaintiff’s counsel in argument contends that the patent as to the spacing of the inlets is not limited as to their exact location or number and points out that two of the forward inlets as originally constructed in the plaintiff’s lehr at the Swindell plant have been boxed because not needed in actual operation for Swindell’s particular type of glassware. But on the whole I am satisfied from the testimony and argument that it is at least true that the primary function in practice of the plaintiff’s inlets is to effect and control the temperature after the annealing stage has been concluded, although it is quite possible that the inlets may at times for particular types of glassware be used to some extent to have effect within the annealing range. Control of the temperature gradient within the second annealing stage in the plaintiff’s lehr is effected principally by the operation of the air inlet located just in the rear of the burners in the receiving end of the lehr, and by a suction fan at the delivery end of the lehr which pulls the hot air through the flue and discharges it there into the outer air. In effect, therefore, the function of the inlets and outlets of the respective structures are for a different primary or principal purpose. (3). But possibly even more conclusive as to non-infringement is the fact that it is clearly proved from the prior art that the defendant’s outlets are old in the art and differ in no material principle in their function in the alleged infringing structure from similar dampered outlets in common practice long before the plaintiff’s patent application. For a convenient summary of the more important evidence on this point, and as illustrating the testimony of some of the witnesses, reference should be made to defendant’s Exhibit 18 which, in Fig. 1, shows a diagram of the more important features of the Amco-Swindell lehr particularly with respect to the function of the uptakes from the defendant’s lower flue which, controlled by dampers therein, serve as outlets for or means for exhausting the heated air which, for clearness of illustration is colored red. It is these uptakes or outlets with dampered control which the plaintiff contends constitute substantial equivalents for the cold air inlets of its lower flue. On this exhibit comparison is readily made between the structure and functioning of the defendant’s lehr with the structure and functioning of other lehrs built by the defendant and other manufacturer's and installed for customers in 1920-1923, some at least, more than two years prior to the application for the patent in suit. Thus in Fig. 2 is represented the lehr built by the defendant for. H. C. Frey & Son in 1920; Fig. 3 that built and installed by the Dixon Company for the Capstan Glass Company in 1921; Fig. 4 that- built for the Hemingray Glass Company in 1920 by the defendant; Fig. 5 that built for the Owens Bottle Company of Charleston, W. Va., in 1923 by the Dixon Company. In the respect referred to, I find no substantial difference in the principle of operation of the defendant’s present lehr and the others just mentioned. It is true there are some slight differences in the details of the structures, as for instance in the comparative length of the upper and lower fire boxes and flues connecting therewith, and in the number and spacing of the dampered outlets, but the principle of exhaustion or subtraction of the hot gases from the heated flues by the dampered outlets at spaced intervals, is the same in all; and the purpose of the spaced dampered outlets was likewise the same in all, to measurably control at intervals the heat gradient in the annealing stage of the tunnel of the lehr. It is true also that in these earlier lehrs, as compared with defendant’s present structure, the uptakes from the lower flues took the form oi definite vertical flues in which the dampered controls operate, while in the defendant’s present structure the uptakes are not flues as such but vertical pipes which, passing vertically upright through the tunnel and not entirely insulated, necessarily have measurably the same effect as vertical flues even though not primarily designed for that purpose but rather as outlets for the heated gases. The Dixon lehrs were made in accordance with the Milner patent No.-1,361,604, applied for April 1, 1919, and issued December 7, 1920. This is one, but by no means the only one, of the numerous patents revealed by the prior art which were reviewed by the witness Schwartz. Special reference may be made to some of the features disclosed in its specifications. Thus the patentee says (page 1, line 38) : “One object of our invention is to provide a leer of this character with an improved flue arrangement for distributing and circulating the heating medium around the annealing chamber, whereby a substantially uniform annealing heat may be obtained within said chamber at any given cross-section thereof, while the heat at different portions of said chamber may be regulated and controlled effectively as may be desired.” And again on page 2, line 42: “As will be seen from Figs. 1 and 3, there is a separate offtake flue 21 for each of the flues 20. Each of these off-take flues is provided with' its own controlling damper 23 (see Fig. 9). In this manner the circulation through the individual flues 20 can be separately controlled and the temperature within this portion of the leer chamber accurately regulated and controlled. The stack flue 22 is common to all the offtake flues 21.” In lines 125, etc., it is said: “The advantages of our invention will be readily understood by those familiar with glass-annealing leers. It provides a leer in which the products of. combustion are kept entirely out of contact with the glassware being annealed; in which practically uniform quiescent heat conditions can be maintained in any given cross section of the leer proper, owing to the manner in which the heating medium is distributed and controlled in the individual flues, and which provides a construction which can be readily built and which greatly facilitates repairs with a minimum loss of time.” Plaintiff’s counsel contends that the defendant’s method of partially regulating temperature in the lehr is an infringement under the doctrine of substantial equivalents. But as already pointed out, the plaintiff’s patent necessarily has a very limited scope in view of the gradual development of the art and cannot validly be made to cover old methods of temperature control which were previously known and practiced in the art of building lehrs for annealing glassware. In this respect the question of infringement is not unlike the situation presented in a case some years ago in this court— Chisholm-Ryder Co. v. Buck, 1 F.Supp. 268, 281, affirmed (C.C.A.) 65 F.(2d) 735—where it was said: “If the defendant is able to apply the principles of a machine which is old in the art to the new use * * * I see no principle of patent law which prevents him from doing so even though he substantially achieves a result which can be better and more efficiently accomplished by the more elaborate mechanical invention of the plaintiff.” Second: As to the Ingle Patent, 1,583,-046, May 4, 1926. Although this patent refers generally to an improved type of lehr for annealing glassware, its special feature has relation particularly to the conveyor belt. In the specifications it is pointed out that generally the conveyors in lehrs have embodied the use of solid pans for holding the glassware; and that this practice had the disadvantage of a comparatively large mass of metal with high heat capacity requiring large fuel consumption, and also retarded the gradual fall of the temperature gradient by retaining the heat throughout its journey, in addition to which it was subject to warping when exposed to changes in temperature which distorted the surface and led to breakage by the fall of the glassware. To overcome these defects the patentee provided a conveyor belt of low heat capacity, preferably of open work structure, such as woven wire fabric which, by viitue of its light weight, reduced fuel consumption and more readily adjusted its temperature to that of its environment and thus contributed to the maintenance of a uniformly declining temperature gradient. But the even more particular feature of the patent was the construction whereby the open work light weight ware-bearing strand of the conveyor belt was made to slide directly on a smooth flat floor of the lehr tunnel and the return strand was not led back within the tunnel but below the tunnel. In considering this patent it is to be noted that it was applied for on the same day, March 31, 1925, on which application was made for the original of the reissue patent just above considered. Mulholland and Ingle were both employees of the Hartford-Empire Company. The two patents, therefore, have a very close relation. In fact the Ingle patent in describing the lehr described quite precisely the lehr of the reissue patent and the similarity between the drawings of the Ingle patent and the related drawings of the reissue patent will be at once apparent. Ingle’s conveyor belt was that of the Mulholland reissue patent and is the form of conveyor belt actually used in commercial practice by the Hartford-Empire Company. It is the “conveyor of low heat mass movable through said tunnel for transporting the ware therethrough” referred to in claim 52 of the reissue patent heretofore mentioned. The Ingle patent as issued contained 13 claims but only one claim, No. 5, is here in suit. It reads as follows: • “A glass annealing leer comprising a tunnel, an endless carrier extending longitudinally therethrough, said carrier being of flexible woven wire construction and having symmetrical upper and lower surfaces, and a drum for driving said carrier arranged to have driving contact with the ware-sup porting surface of said carrier." (The important phraseology has been italicized.) It will also be noted that this claim in suit does not contain as an element the particular feature of the patent specifications, that is, the construction whereby only the ware-bearing or upper strand of the conveyor belt is within the tunnel of the lehr. All the other claims do-contain this element. Obviously the limitation of the suit to claim 5 is because the conveyor belt used in the defendant’s-structure does not involve this feature, its construction being that ordinary and usual to glass lehrs, that is, both strands being contained within the tunnel. An examination of the file wrapper of the patent also shows that the broad claims originally submitted by the patentee for a conveyor belt of flexible or woven wire structure were cancelled after rejection by the Examiner on reference to the prior art. In addition to the disclosures in the prior art referred to by the Examiner, Schwartz, defendant’s expert in this case, has also reviewed the,subject more extensively. It appears that woven wire conveyors had for 50 years or more been used in various types of ovens for baking and for annealing glass. Tondeur, No. 303,195, August 5, 1884, claims use of a woven wire cloth or apron in conveyor belts for annealing plate glass; Meyer, 564,-568, July 21, 1896, discloses its use for the bottom work of pans to hold glassware in the annealing process, and points out many of the same advantages in the use thereof that were referred to by Ingle; Jacobson, 649,695, May 15, 1900, shows the use of woven wire conveyor belts in a baking oven; and Swinscoe, 712,212, October 28, 1902, shows the use of wire fabric for belts or aprons generally. And two German patents, one to Strandh, 224,-451, and another to Schwartz, 237,991, likewise show various types of woven wire belts used in various stages of glass making. It seems impossible to successfully contend that there was any novelty in the use of a woven wire fabric conveyor belt for .glass annealing; and indeed Ingle does not insist in his specifications on woven wire although he expresses a preference therefor. The only other distinguishing element of claim 5 is the requirement that the woven wire conveyor be driven by a drum which has driving contact “with the ware-supporting surface of said carrier.” There is no mechanical novelty in the use of a drum for driving a belt; and neither the specifications nor the file wrapper of the patent, nor the evidence in this case, indicates any reason for the particularity of construction whereby the ware-bearing surface of the woven wire belt should have driving contact with the drum, rather than the other surface of the belt. But presumably, as the file wrapper shows that the flexible woven wire construction of the belt was not of itself a patentable feature, the grant of the claim must have been based on the combination of the two features. The testimony in the case discloses no satisfactory reason for one mechanical arrangement rather than the other, that is, whether the inside or outside of the ware-bearing strand should be in direct contact with the drum which drives the belt. It is therefore somewhat difficult to understand the basis of the grant of this claim by the Patent Office; but it seems clear enough, to save the claim from invalidity by reason of the prior art, it must be construed as having a very narrow and literal import. And on the issue of infringement, it is to be noted that the defendant’s drum drive with respect to this element of the patent is the exact opposite. That is to say, in the defendant’s apparatus the drum comes in contact not with the ware-supporting surface of the belt but with the other side. Literally therefore the claim does not read on the defendant’s structure. Furthermore, the defendant’s driving mechanism from the conveyor belt is precisely in accordance with long prior practice in the art. Gilman, 840,815, January 8, 1907, shows a conveyor apparatus for grain with drive arrangement similar to that in defendant’s structure. The Tondeur patent more than 50 years ago disclosed the same driving principle as that now used by the defendant. As to this, the expert Schwartz said: “Now, it seems to me that the early Tondeur construction used the same principles of driving exactly as are found in the defendant’s structure, because they both drive on the inside of the belt by friction. They both have woven wire belts, which is one of those listed in the patent. They both travel in the tunnel which anneals glassware, they travel over rollers. It seems to me that meets all the specifications in this alleged invention.” As there is no difference in the result between the inside and outside drive, and both perform substantially the same function, it is clear enough that the defendant’s method of driving the belt although the opposite of the' plaintiff’s, would be only a substantial equivalent thereof if the plaintiff had been a pioneer in the use of a drum drive for conveyor belts in glass lehrs or for analogous uses. It is apparent, however, from the prior art that Ingle was in no such position and it seems impossible to sustain his patent at all except when strictly and narrowly limited to the precise and particular arrangement which is claimed, and which the defendant cannot be held to have infringed when he uses a mechanical device which has been known and practiced for similar purposes lor many years. The plaintiff does not claim the right to exclude others from the use of the woven wire belt except in combination with the drum drive; and there is no novelty in the latter, and hence none in the combination which achieves no new result not attributable to the virtue of the woven wire fabric for the belt. The purpose of the conveyor belt is only to transport the glassware. Woven wire does this no better than other materials. Its only advantage is that of open work fabric which has low heat capacity and thus tends to save fuel and facilitate heat transference. The advantage of open work or perforated conveyor belts had previously been pointed out in earlier patents. It seems to be true that the plaintiff first adopted woven wire fabric for the conveyor belt in the modern type of glass lehr, but it had been long ago used in other types of lehrs, and in other analogous arts, and the claim for its exclusive use by the plaintiff was properly rejected by the Patent Office. If the plaintiff is