Daylight Glass Mfg. Co. v. American Prismatic Light Co.

142 F. 454, 73 C.C.A. 570, 1905 U.S. App. LEXIS 4128

• Court: Court of Appeals for the Third Circuit
• Decided: December 11, 1905
• Docket: No. 44
• Status: Published

Opinion

BUFFINGTON, District Judge.

This case involves two appeals from a decree entered in the Circuit Court for the District of New Jersey (140 Fed. 174). In that court a bill was filed by the Daylight Glass Manufacturing Company against the American Prismatic Light Company, alleging infringement of four patents owned by it, numbered, respectively, 695,282, 695,283, 695,284, and 710,434, and which were granted to G. K. Cummings. That court entered a decree adjudging patent No. 695,282 valid, finding infringement thereof by respondent, enjoining such infringement, and dismissing the bill as to the other three patents. From such decree both parties have taken appeals to this court.

As the three last recited patents are all based on the same basic or generic act involved in the machine patent No. 695,282, all four, in the view we take of this case, may be jointly considered and disposed of under a discussion of the machine patent noted. That patent, No. 695,282, was for a machine for making prismatic glass, was applied for May 18, 1898, and was granted March 11, 1902. The specification recited that the prior general practice was to make prism glass by a molding process, and that no other practicable means or device was known or used. The patentee then averred:

“My improved machine is provided with a revolving roller, a supporting device to hold the glass against the roller, the said parts having a traversing motion relatively to each other, and one of said parts being provided with parallel ribs of a prismatic form corresponding to the depressions to be made in the glass.”

This language is carried into the first, second, and third claims, which alone were herein questioned. The elements of such first claim are: “In a machine for making plates of prismatic glass” the combination of, first, a revolving roller; second, a supporting device to hold the glass against the 'roller; third, the roller and supporting device shall have “a traversing motion relatively to each other”; fourth, one of them is provided with parallel ribs of prismatic form; fifth, these ribs run parallel to the direction of the traverse movement of the roller.

Validity of the claim in view of the prior art is denied. In considering the question of the patentable character of the machine in question, we must not be misled by the fact that its use has been attended with commercial success in the way of a large, better, and cheaper product; for, in the steady advance incident to progress in manufacturing, many nonpatentable processes and methods have proved most original and exceedingly profitable, and it must be remembered that everything novel and useful is not therefore necessarily patentable. Also, in taking up the question of the patentability of Cummings’ roller table, we must charge him with knowledge of all that preceded him in the art, for “it is a presumption of law that all mechanics interested in upholding or defeating a patent were fully acquainted with the state of the art when they took out their patent, or when they built their machine. * * * Each party may then be assumed to have borrowed from the other whatever was actually first invented and used by the other.” Crompton v. Knowles (C. C.) 7 Fed. 199; and Mast, Foos & Co. v. Stover Mfg. Co., 177 U. S. 493, 20 Sup. Ct. 711, 44 L. Ed. 856, where it was said:

“Having all these various devices before him, and whatever the facts may have been, he is chargeable with a knowledge of all pre-existing devices, did it involve an exercise of the inventive faculty to employ the same combination in a windmill for the purpose of converting a rotary into a reciprocating motion.”

Now this patent relates to the rolling of prism glass. Such glass had been used for many years for lighting interiors by means of refracted rays. This was done by making on one side of the glass, prisms of triangular cross-sections, which serve to deflect or refract the light to any desired point. The art, then, of prism lighting, was old and its principles known. The method of making prism dies of triangular cross-section in the bed or bottom of a metal mold, so that melted glass, when poured thereon, would form light prism, was well understood. The art, too, of rolling several kinds of glass, as contrasted with pressing it in molds, was also old; its advantages in dispensing with skilled labor well known; and its practice, and the mechanical agents necessary to accomplish it, were familiar to glass-makers. The general knowledge of that practice may be well illustrated by patent No. 487,803, of 1892 to Walsh for a machine for rolling corrugations on both sides of plate glass sheets. The general mechanism of a bed-plate on which was mounted a traversing roller (b'oth being of a type which the proofs show had been employed by Danenhoffer for some 20 years), was used by Walsh. In his machine we find the first three elements of the claims we are considering, viz.: First, a revolving roller; second, a supporting device to hold the glass against the roller; and, third, the roller and supporting device have a traversing motion relatively to each other. Now, on both the roller and the bed-plate, Walsh provided ribs of semicircular form, both parallel to the direction of the travel of the roller. He says:

“Figure 1 is a side elevation of a glass-rolling machine in which the roller travels along the length of the bed-plate of the machine. Fig. 2 is a plan view of the same, showing the bed-plate having continuous uninterrupted parallel channels or corrugated longitudinally and the roller circumferentially.”

These roller and bed-plate grooves, it will be seen, answer precisely to those of the fourth element of Cummings’ claim, save that Cummings’ ribs are of a prismatic, while Walsh’s are of semicircular cross-section. Moreover, as stated in the fifth element of Cummings’ recited claim, these ribs run parallel to the direction of the traverse movement of the roller, just as do Walsh’s. It will thus be seen that the only difference between Walsh’s and Cummings’ rollers consisted in the latter substituting a prismatic or triangular rib for a semicircular one, on the roller or bed-plate, optionally, leaving the other plain. Indeed, on citation of this patent in the office, the claims of Cummings were at first rejected, the examiner then holding:

“It is no invention to shape the ribs to the shape of the prism desired in the glass.”

In so holding, it would seem he was in accord with authorities of weight. In Stimpson v. Woodman, 10 Wall. 117, 19 L. Ed. 866, a machine was patented for giving a pebbled surface to leather. Pebbled leather as a product, made by a hand tool and also made by a hand-operated pebbled roller, were old, as was also a machine with a plain roller. It was held there was no invention in die-pebbling a roller and placing it in a machine. In the same line is Smith v. American Bridge Company, 3 Barr & Ar. 565, Fed. Cas. No. 13,002, where it was held:

“It being old to cut dies on the face of a trip-hammer, or on the anvil on which the trip-hammer works, so as to forge iron into different shapes, there was no invention in making dies in such hammers or anvils of the proper shape to make chord-heads for iron bridges.”

In Peters v. Active Mfg. Co., 130 U. S. 626, 9 Sup. Ct. 643, 32 L. Ed. 1057, it was said:

“The whole of this alleged invention is based upon the idea, old and well known, that a metallic die, whether of a cameo or intaglio form, will, when impressed upon a piece of heated or yielding metal, leave the latter of the converse form of the die, and that, when two dies are brought together over a piece of heated or yielding metal, the latter will take the shape of the space existing between the contours of the two dies. It is an inevitable consequence of the use of two dies in such a way, on two pieces of metal of proper size, heated to a welding heat, that swaging or welding will take place by the impact of the dies; that, when the dies have tongues and depressions in them, the metal acted on by such tongues and depressions will take the shape, in form and thickness, of the space left between the tongues or the depressions; and that a greater or less thickness of metal will be the result as the face of the tongues is more or less inclined. All this was old and common knowledge, and the whole of the operation resulting from such features is nothing but the well-known action of two dies so shaped as to give the desired conformation to the article acted on by them.”

Indeed, a study of the voluminous record in this case convinces us that resort to the inventive faculty was not necessary in the construction of the rolling table here in question. It was a purely mechanical process in conception and execution. When the patentee wanted to roll glass he employed a rolling table to make it by putting grooves on the roll in precisely the same way Walsh had used it to make straightaway, corrugated grooves. For while the proofs show that rolled prism glass was desired in that business and its value recognized, the reason1 why it was not made was not because it was not known that prism glass could be formed on glass by rolling, but for the wholly different reason that it was thought and believed that if made the glass could be neither annealed when rolled, or cut when annealed. While this assumption was wholly false, yet it is equally true that it existed. This mistake, and not inability to roll prism glass, prevented the adoption of rolling. We think this is clear from the proofs, which show quite conclusively that owing to what are known as the “strains and stresses” imparted to prism glass in molding, they cannot be cut to predetermined shapes. It was therefore argued, if you cannot cut a tile prism of three inches square, what advantage is it to have a rolled sheet three feet square? These difficulties of cutting and annealing, and not the rolling difficulties, were the obstacles.in the path of advance in this art. This is clearly shown by complainant’s witnesses. Thus Mr. Wright says in answer to the question:

“Had rolled sheet prism glass ever been upon the market or in use, to your knowledge, prior to the time when Mr. Cummings, or his company, put it on the market as described in the last answer? A. No, sir; it had not, although it had been suggested by people in the glass trade as well as architects and builders, and it was recognized that if prism glass could be rolled in large thin sheets it would at once create a largely increased demand, and would be used in a great many buildings in which the molded tile would not be used on account of the price. The general opinion, and, in fact, the statements of practical glass manufacturers and men in the prism business, was that it would be impossible to roll a large thin sheet of prism glass, which would have sufficient strength to resist strong wind pressure, and which could be cut into any desired size or form, and not until after the glass was actually made and its strength tested by scientific methods were these opinions shown to be wrong. Q. How about the attitude or opinion of the trade at that time in reference to the practical feasibility of cutting the rolled sheet prism glass, if it could be made in sheets? A. Prior to the introduction of rolled sheet prism glass, the molded tiles had been on the market for several years, and were being used by four prism companies throughout the United States. It was the unanimous opinion of all the prism companies and the manufacturers of the molded tile that, if the rolled sheet prism glass was made according to Mr. Cummings’ application, it could not be cut successfully. This opinion was arrived at after several years of experience in trying to cut the molded tile. They argued that, if it was impossible to cut the molded tile commercially, it would be infinitely more difficult to cut a sheet of prism glass, which would be 10 times as wide and 20 times as long.”

So Mr. Cummings, the patentee, in discussing the desired uniform thickness of the corrugated glass rolled by the Walsh machine, says:

“Every indication being that Mr. Walsh considered, just like every, one else in the art at that time, that rolled sheet prism glass, having a particularly varied cross-section, could not be possibly ‘effectively annealed.’ ”

His testimony was that Mr. Heidt, a machine builder to whom he went to build a prism glass roller, was under this impression. Thus:

“He said that it was impossible to make prismatic glass by rolling it; that the resultant sheet prism glass could not be disengaged from the bed-plate or from the roller without either breaking the prism edges off, and that they would then remain in the bed-plate or the roller, and have to be chipped out; that, even if this were not so, in disengaging the rolled glass the prism edges would be mangled and distorted. Ete also said that, providing the glass was made and disengaged from the roller and bed-plate, in the leers it could not be properly annealed, and, in his opinion, a great portion of it would crack or fly to pieces. He further said that, even if it were made and properly ennealed, it could not be cut with any degree of practicability, especially in reference to any cut that would run in curved lines, or transverse to the lines of the prism, except at a perfect right angle. In fact, he told me that I spelt ‘failure’ before I started. It was not, however, until the end of this conversation, that it occurred to me that perhaps he did not understand my exact process, and, when I explained to him the exact principle of running the prism lines exactly parallel to the motion of the roller and bed-plate, he modified his statement to the extent that it might be possible to be done in that way, and only in that way; but still he didn’t believe it.”

He also quotes others as under this, as it afterwards turned out, erroneous view as to cutting, viz:

“Also J. Dannenhoffer, a manufacturer of glass in Brooklyn, also Mr. Dougherty’s superintendent, and they all agreed that rolling prismatic glass according to my invention was impracticable, and that, even if it were made, it would be a very imperfect product, that it could not be cut, and that it would be structurally too weak to withstand extreme variations of temperature.”

This assumption by the glass manufacturers that sheets could not be cut and that it would be impossible to anneal them was quite natural, in view of their experience with molded glass, and they had every ground to reason by analogy from that assumption it was no use to roll large sheets, when they had to make the molded tiles thick and small in order to anneal them, and that they could not be satisfactorily cut. Now the rolling of glass has shown this assumption was wholly wrong, that rolled glass can, when made thin, be successfully moved to the leers, can be annealed in large sheets, and that, in cutting, it acts differently from molded tiles. But these facts do not impart inventive qualities to the making of a machine to roll such glass. That there were mechanical difficulties to be overcome in successfully rolling prism glass goes without saying. Indeed, the proofs in this case show that Cummings’ preferred method of putting the grooves on the roller did not prove successful and that it was eventually superseded by the other suggested method of putting them on the bed-plate. But to our mind all these steps, important as they were to the art, were mechanical, and not inventive in their nature.

Moreover, the teachings of the art were not confined to the Walsh patent already noted. To our mind there were a number of other patents which threw much light on the rolling of prism glass. Under ordinary circumstances, a patent which caused no ‘advance in its art should have small effect upon a successful patent of 20 years later; but having seen that all use of machine for rolling prism glass was prevented by the mistaken estimate of its worth, the teachings of an earlier patent, which did not result in making machines, cannot for that reason alone be disregarded. Among these is the patent to Stevens, No. 3,549, of 1869. In it we find a machine for rolling corrugated glass, composed of a roller and a bed-plate of the general type shown in the Walsh and Cummings devices, and embracing the three elements of Cummings’ first claim, viz: First, a revolving rollert second, a supporting device to hold the glass against the roller; and, third, the roller and supporting device have a traversing motion relatively to each other. The patent states:

“The form of the glass is determined by the formation of the face of the pressure roller which presses the glass against it. * * * Corrugated sheets of glass may be produced by the use of a traveling roller and fixed table, or vice versa; the roller and table having corrugations to match each, other. * * * In producing corrugated sheets of glass by rolling and pressure, both the mold bed, B, and roller, C, are grooved as shown in the drawings. * * * The glass to be rolled is poured upon the mold bed in the plastie state known as ‘metal,’ and the bed is then transversed beneath the roller, C, by turning the handle, G, whereupon the metal will be rolled out or spread over the surface of the mold bed, and will assume whatever form has been imparted to the roller and bed, or to the roller or bed, as found most suitable-in practice.”

The drawings show the mold grooves run the length of the bed, and consequently the grooves on the roller, the roller and table having corrugations to match each other, were circumferential. It will thus be seen that at this early day this patent discloses a machine embodying every element of the Cummings claim here involved, save that the patterns were corrugated ribs, instead of triangular prisms. This was followed in 1880 by the patent of Boughton, No. 139. The-object of this patent was to brighten interiors by rays refracted through prism lights and to manufacture the same. Boughton’s prism consists of a number of parallel projecting ridges, resembling ratchet, teeth, on one side of the glass; one surface of the ridge being nearly at right angles to the sheet and the other inclined. Boughton describes his method of rolling his prism sheets as follows:

“Tbe sheets of glass are made in the ordinary manner of making rolled! sheet glass. The molten glass is poured onto a bed, the surface of which has been planed or shaped to leave upon it a number of parallel projections to correspond with the depressions it is required to make in, the face of the-sheet of glass. The glass is poured onto the bed in front of a heavy roller, which is then caused to roll over the bed and spread the glass uniformly over it; the surface of the roller being kept at the required distance from the bed by strips or ‘fillets laid upon the bed along its sides, and of thickness equal to-the thickness of the sheet of glass to be made. In place of the projections-being upon the bed-plate the bed-plate is sometimes smooth and the projections are formed on the roller.”

Because Boughton does not state that the prism molds in the bed-plate run the length of the bed-plate or those in the grooves are circumferential, it is contended the patent does not anticipate Cummings But the question here is not one strictly of anticipation, for it is not sought to show that a patent, otherwise valid, is anticipated by a substantially similar structure; but the question is, in view of the existence of such a machine as Boughton’s and the teachings of his patent,, did it involve invention to make Cummings’? Now, while Bough-ton does not state his prism patterns were to run with the table around the roll, he does not state that they were to be transverse. Consequently,, there is no more ground for saying they are crosswise than that they are lengthwise; and, indeed, it would seem the natural mechanical way, not only to a glass maker, but a roll turner as well, was to make the roll grooves circumferential, rather than transverse. The prior patent of Stevens showed circumferential grooves, and a circumferential groove is not only easier shaped than a transverse one, but the general practice of the rolling art would point toward circumferential grooves on the roller of a rolling machine, or to sink longitudinal ones in its bed-plate. The facts in this case bring it within the principles of Blake v. San Francisco, 113 U. S. 679, 5 Sup. Ct. 692, 28 L. Ed. 1070, and Penna. R. R. Co. v. Locomotive Truck Co., 110 U. S. 490, 4 Sup. Ct. 220, 28 L. Ed. 222, wherein it was held that:

“The application of an old process or machine to a similar or analagous subject, with no change in the manner of application and no result substantially distinct in its nature, will not sustain a patent, even if the new form of result has not been heretofore contemplated.”

We are accordingly of opinion the machine patent No. 695,282 was void for want of invention, and that the part of the decree of the Circuit Court sustaining it was erroneous; and from what has been said it also follows that the portion of the decree of the lower court adjudging Nos. 695,283, 695,284, and 710,434 to be invalid was right.

The order of this court, therefore, is that the decree of the Circuit Court be reversed, and the cause remanded to that court, with direction to dismiss the bill of complaint, with costs.