Schmertz Wire-Glass Co. v. Pittsburgh Plate-Glass Co.

BUFFINGTON, Circuit Judge.

This case concerns a comparatively new article of commerce called “wire-glass,” which is a sheet of plate glass having imbedded in its center a fine wire-web. This web strengthens the sheet and prevents its being shattered by fire or a blow. Rough wire-glass is cast in two varieties, one adapted for polishing, the other not. Unpolished wire-glass is used for skylight and area purposes. Polished, or wire-glass plate, is used as ordinary plate glass in large buildings for window lights. In addition to window-light service, however, wire-glass plate, by reason of its imbedded mesh, serves the further purpose of an iron shutter. It is thus a protection against entry and a fire retardant. A just estimate of the radical difference between rough, unground, sheet glass and polished plate glass recognizes the qualities which permit the one and forbid the other to be polished. One is limited to coarser use and smaller price, while the other may be applied to higher and wider uses and commands better prices. An appreciation of these important manufacturing and commercial factors is essential to a just estimate of the patents here involved. It follows, therefore, that the use of the generic term “wire-glass” as alike applicable to the development of both rough, but unpolishable, wire-glass, and rough, but polishable, glass is misleading. All wire-glass is, when it comes from the casting table, rough glass. The rough glass proper ends its development at that point and passes to the uses to which it is limited. The other, while it is of course adapted'to like uses, has the higher possibility of being ground and polished and thus reaching the condition of polished plate, with the superadded qualities which flow from the imbedded netting. To displace blown glass in window lighting, the plate-glass manufacturer was forced to produce a very thin sheet of plate. This was imperative to get light-weight windows to raise, and also for economy of space and cost in counterweighting. The extra weight of plate over blown glass virtually forbade the use of cheap iron balances and necessitated more expensive lead ones. These factors of cost and space determined the choice of plate or blown glass in buildings, and, as the height of buildings and window space grew, plate thinness became a greater factor in widening the use of plate. This condition was met by the production of very thin plates, and as a result they have largely superseded blown glass in both private and public buildings where the use of the heavy plate of the old art would have been impossible. But with this use of a light, thin plate in modern tall buildings a greater danger from fire became apparent, for these light sheets were shattered by fire; and pieces, falling from heights, made them very dangerous to firemen. It therefore became apparent, after rough wire-glass was successfully made, that, if the wire mesh could *75l e so regularly and centrally imbedded in a thin sheet as not to prevent its being ground and polished, a new and valuable factor of utility and safety could be embodied in modern construction. But it was equally dear that, if at any point of the sheet any single part of the wire came near either surface so that its strand would be bared by grinding, the 'beet could not be polished.

Now the first thought in connection with this art is that, when the idea of making wire-glass is once suggested, the manufacture must be very simple; in other words, that the inventive thought, if one exists, consists in the conception of wire-glass. Casting tables and rollers for smoothing glass are old. The casting of sheets of rough glass is obviously simple, and grinding and polishing a developed process. What, then, more simple than to cast a sheet, then lay a web of wire netting upon it, and on the two pour and cast a second sheet? And that such should be the first impression is quite natural, for this was the idea and practice for years of the whole glass art, and it found expression in the patent development of three foreign countries and in the name “European process.” But experience, as we shall see, proved that this seemingly simple method did not and could not produce wire-glass, and that even rough wire-glass was only a hoped-for product, until, in 1892, Shuman, an experienced American glassman, solved the rough-glass part of the wire-glass problem in a totally different way. And as Shuman’s invention was a step in the development of rough, unpolishable wire-glass, which culminated in Sehmertz’s polisliable wire-glass and in the new commercial article of wire plate, it is necessary, in order to understand Schmertz’s final development, that Shuman’s partial development be clearly grasped. In his device Shuman discarded the theory of sandwiching the wire netting between two sheets cast at different times, or, indeed, of using rwo sheets at all. He conceived the idea of forcing the web into the body of a single sheet and then closing up the scars caused by this insertion. His machine was a long table mounted on and heated by a furnace and provided with three rollers above it. Molten glass was poured on this hot table and pressed to the desired thickness of the sheet by a front, plain-faced roll. A second or rib roll, engaging the netting delivered in a chute back of the first roll, then pressed such netting, in a waving, uneven line, into the body of the sheet. The third roll, which was plain-faced, followed and covered the openings nade by the insertion of the wire. Shuman’s patents for machine and process were sustained in the Circuit Court of Appeals, Seventh Circuit, in Streator Cathedral Glass Co. v. Wire-Glass Co., 97 Fed. 950, 38 C. C. A. 573, Mr. Justice Brown, of the Supreme Court, sitting and delivering the opinion. It was there said Shuman’s process involved “not merely the function of a mechanical device, but certain elemental action. * * * It is, in fact, a series of acts performed with molten glass and wire-gauze, by which they are transformed into a separate manufacture, within the definition of a process patent in Cochrane v. Deener, 94 U. S. 780, 24 L. Ed. 139”; and if was held the proper subject of a process patent. The relation of Shuman’s *76invention to the practical commercial origin of wire-glass was stated by Justice Brown:

“The Shuman patent was evidently the first practicable method of making wire-glass, and appears to have attracted a good deal of attention in this and other countries, and various medals were awarded to the inventor.”

Judge Jenkins, in a concurring opinion, says:

“Others * * * had conceived the thought of wire-glass, and provided certain means for the manufacture; but they never achieved success, or pointed out any practical means for its accomplishment. There was no wire-glass in use or upon the market, either in Europe or in this country, when Shu-man obtained his patent. Charging him, as we must, with knowledge of the prior art, he was not the first to conceive the idea of wire-glass, but was the first to make it possible.”

These judicial findings that the prior wire-glass art was theory and had produced no practical results restate the estimate of the commercial and scientific world. The Franklin Institute of Philadelphia, in awarding a medal to Shuman, in the report of its committee said (volume 187, No. 819) :

“In conclusion, they find that the process and machine of Mr. Shuman not only represents a marked advance upon all previous efforts in the class of Inventions to which they relate, but also that they meet for the 6 time, in this branch of the arts, the requirements of a practically operative method and means of manufacture.”

The'Iron Age,,November 24, 1892, says:

“The idea of making a strong combination of wire and glass is said to have had its inception some years ago in England, but it never seems to have gone beyond the region of experiment, although some small panes, not more than one foot square, of an imperfect kind, were made, but the material never obtained recognition as an article of commerce on account of the great cost of production and limitation in the size of the sheets. It remained for an American to carry out the idea to a practical issue. Prank Shuman, after patient and exhaustive experiments, has overcome the initial difficulties, and is now able to make sheets of wire-glass by a process of his own, of any desired size or thickness and of perfect consistency, at a cost which will bring it within the reach of all who now use ordinary glass.”

And to the same effect are the Scientific American, November 5, 1892; Paris Revue Industrielle, April 8, 1893; Encyclopedia Britannica, vol. 5, p. 3173.1

Now the contention of the complainant is that Schmertz, the patentee in suit, was the originator of polishable wire-glass, as .Shu-*77man had been of rough, ribbed, unpolishable wire-glass. On the other hand, it is claimed the Schmertz patents are void by reason of the disclosures of the art prior to Shuman. In support of this contention there are cited 20 British, 4 French, 4 German, and 41 American patents; 3 publications alleged to disclose the art, and 13 prior practices thereof. These have all been examined, and the patents were each discussed in detail as a part of this opinion. In view of the facts that, as stated above, both in judicial opinions in which we concur, and in contemporaneous literature, the wire-glass art had its first practical development in Shuman, we have omitted a discussion of the prior art which we made and but" for the length of this opinion had purposed to insert. We content ourselves with saying the number of references and citations prove almost too much. It is simply incredible that if Schmertz’s process was disclosed by any or all of these references covering, as they do, a period of almost 50 years, the commercial production of wire-glass would not have followed. Indeed, the very fact that there was so much literature and inventive effort in the art without.practical result is persuasive that its ultimate solution was invention of a high order. The truth is that, plausible as prior patent disclosures often seem, and useful as they are to accurately define and justly limit the area of subsequent inventive effort, it may be accepted as a general truth that if they resulted in no practical development they lacked some element which the successful patent they are alleged to anticipate possessed. Accepting, therefore, Shuman’s process as the then zenith of the wire-glass art, it was found that, meritorious as it was in producing rough wire-glass, it was open to objections which, when Schmertz disclosed his invention, led to the total abandonment of Shuman’s process. The defects of Shuman were that his process was a one-sheet pour and the web was imbedded by fracturing that sheet. As we have seen, the first roll flattened the pour into a single sheet. As this came from the roll, it was exposed to the air and the rapid hardening of a surface unprotected by a second pour. This produced what is known in the glass art as a skin. Now this skin was ruptured both by the ribs of the second roll and by the meshes of the web as it was forced through it. The scars thus made were not effaced by the third roll. A glass skin when once ruptured cannot be restored in clearness, and Shuman’s process, therefore, did not produce clear, unmarked glass suitable for polishing. It showed longitudinal lines made by the fracture of the skin by the roll-ribs and also cross-markings made by the wire mesh. These gave to Shuman’s product the name “Caramel” glass. The chill to which the single sheet of the Shuman process was subjected in its treatment by-three rolls restricted its output to sheets of about eight feet. In longer ones the third roll could not clear out the hardened lumps and roughness made by the mesh and rib-roll.

In this state of the art Schmertz took up the problem of making a wire-glass suitable for polishing, and applied for two patents, each containing process and apparatus claims. In his specifications, which are practically identical, Schmertz cites the Shuman process as the then recognized method of making wire-glass. His language is singu*78larly clear in stating the defects of Shuman’s process and the exact object he himself had in view. He says:

“The purposes of my invention, generally stated, are to devise a certain combination of apparatus and such changes in the process of manufacturing wire-glass as will produce a glass with brilliant top surface instead of rough and dull; also, to prevent the glass becoming overchilled in making, so that it will not be excessively hard. Wire-glass as it is now actually manufactured in this country is made in the following manner: The glass is poured upon the usual casting-bed and a roller pulled over the same, developing a sheet of glass of substantially the thickness of the final product. Back of this roller, which is sometimes termed the ‘smoothing-roller,’ wire-gauze is fed upon the top of the sheet thus formed from a chute or other suitable device. A second corrugated roller now passes over the wire-gauze and the sheet of glass, and by its ribs deeply indents the wire-gauze in the body of the sheet. A third smooth finishing roller now travels over the sheet thus corrugated and gives a finish to the top of the sheet, gome of the objections to this method are these: The action of the three rollers chills the top surface of the glass excessively before it can be transferred to the annealing-kiln. The result is that the glass becomes unduly hard and is difficult to cut. It is a serious task to cut wire-glass when made under the best conditions, and this extra hardening aggravates the situation; but another and more important objection to the method just described lies in the rough top surface produced. The first or leading roll gives a smooth finish to the top of the glass; but this is immediately ruined by the passage of the corrugated roll in the operation of forcing the wire down into the body of the sheet. This smooth finish is never regained, for although the third roller will efficiently smooth down any elevations on the surface of the glass it will not perfectly transfer the surplus glass to the depressions and fill them up with a perfect weld. An examination, therefore, of glass made by this process will disclose fine channels running across the surface which has been uppermost in the process. These channels roughen the surface, making it appear dull and lusterless, and greatly impair its appearance. The unevenness of the surface again adds to the trouble in cutting the article, as the diamond will not traverse it with the same ease as avpolished surface. I proceed upon a radically different theory of forming the product, and secure a sheet with brilliant top surface, and also one which is as readily cut as is possible with composite articles of this character.”

His apparatus is shown in the accompanying cut taken from patent No. 791,217, that in reissue No. 12,443 on No. 791,216, being the same, save that the wire netting is fed in advance of the first roll. He describes the cut as follows:

“In the accompanying drawings, which make part of this application, 2 is the ordinary casting table, having the usual racks, 33, on its longitudinal edges. *7941 are tJie hand wlieels, which by pinions, 55, advance the leading roll, 6, which may be either smooth or corrugated, and secondary roll, 7. Those rolls rest upon trangs, 88, and preferably one-fourth inch in height, and which determine the thickness of the glass, subject to a special provision in the leading roll hereinafter described. !>9 are the guides which fix the width of the sheet and are pushed ahead by’ the rolls. 10, 10, are the rods connecting the right and left hand guides”—

and states that all the above mechanism is old and well known. He then proceeds:

“11 is a chute for wire, 12. located at the rear of the first roll. The leading roll, 6, has its ends, .13, 13, recessed, preferably one-eighth of an inch, so that the body of the roll is left, one-eighth of an inch above the bed of the table, while the body of the second roll is one-fourth of an inch above the bod.”

The specification then proceeds:

“The operation of the device is as follows: A ladleful or suitable amount of glass is poured upon the table in advance of the leading roll, and the leading roll moved forward. In practice the leading end of the sheet thus formed is discarded as made of chilled glass. Therefore I do not commence to feed the wire down the chute until the first roll has made a. sheet of glass sufficiently long to project about 18 inches back of the wire-chute. At this stage the wire is fed down upon the top of the sheet, and finally lies flat upon the same, as shown in Fig. 5. As the forward or dead end of (he sheet of glass moves near the rear roll, I make a second casting upon said dead end, that the imperfect ends of Hie two sheets may coincide, and thus economize material. The rear roll advancing crushes down this second casting and forces the same over and through the wire-gauze, firmly welding the material in the second easting to the wire-gauze and to the first sheet, and firmly uniting the three component parts. A brilliantly polished top surface is imparted to (he product by the passage of the second roll, and the wire is left substantially in the center of the finished sheet, whose thickness will be the height of the rear roller above the bed of the table. The top sheet having only one roller traversing It will not become so chilled as where two or more rolls pass over it, and especially where one is a corrugated roll throwing the glass up in ridges. My invention is distinguished from tho art as practiced in tills country not only by the superior quality of the product, but as well by the following features, either singly or in combination, viz., by the use of two rolls instead of three or more, by Hie formation of an original sheet a fraction of the desired thickness, feeding the wire upon the top surface of said sheet, and subsequent welding ro both wire and original sheet of a top sheet with unbroken surface formed from a second easting while the first sheet is being completed.”

These extracts taken from patent No. 791,217 are illustrative as well of his reissue No. 12,4-13 of patent No. 791,216, the difference being that the latter has a pre -roll, instead of the post-roll, wire feed of No. 791,217.

It is alleged that Schmertz’s statement of the art was inadequate and misleading. We cannot agree thereto. He rightly described the Shuman process as the only practical prior process, and his statement in that regard accords with the conclusion subsequently reached by Mr. Justice Brown and Judge Jenkins. Moreover, had Schmertz meant to conceal from the Patent Office knowledge of the unsuccessful European method, it would not have availed him, for the French inventor Appert, patent No. 608,096, with whom he was at once involved in interference, called attention both to it and Shuman’s practice, saying:

*80“The importance and value for various purposes of glass reinforced by a metallic trellis or net work embodied therein are well understood and need not here be set forth. It has therefore been very desirable to discover a process which, while admitting of the production of such article under conditions analogous to those attending the manufacture of ordinary rolled glass, and consequently at a moderate price, would preserve all the essential qualities of limpidity and transparency of the glass without reducing its resistance or increasing its fragility. Attempts in this direction were made in 18S6, when it was proposed to place a metallic trellis or network between two glass plates and unite the latter. That attempt, however, was unsuccessful, as the product possessed neither homogeneity nor cohesion. In 1892, it was proposed to obtain the desired result by forcing into the body of molten glass a metallic network or trellis. The chief difficulties attending this process are that a netting of- light wires (such as it is very important to use, so as not to impair the essential properties of transparency and limpidity) could not be forced in that manner into the body of the glass, and that it is not easy by this plan to imbed the netting regularly and at the proper uniform distance from the surfaces of the sheet.”

We take this occasion, now that Appert has been mentioned, to say that a clear apprehension of his patent becomes necessary to a proper construction of the claims of Schmertz’s patents. The statement of the Schmertz reissue, No. 12,443, is:

“Desiring to correct by reissue the accident and mistake whereby claims constituting the issues of a certain interference and covering broadly the invention common to the patents of Schmertz, No. 791,216 and No. 791,217, and the patent of Leon Appert, No. 608,096, were omitted from the specification of said patent No. 791,216, what is claimed is — ”

Both Schmertz and Appert, who were working toward the same end at the same time, took the same process path. Tike Schmertz, Appert discarded the three-roll, one-sheet process of Shuman and adopted a two-roll, two-sheet process, wherein the two sheets were simultaneously formed with the- wire web between them. The principle of simultaneity is clearly brought out by Appert’s specification. Appert states that his process — •

“is characterized by the continuous and progressive formation of two layers of molten glass and the simultaneous introduction between them of the metallic trellis or net work, and the compression of the whole into a solid, uniform, and homogeneous body or sheet. The simultaneous formation of the two layers, with the fabric inclosed between them, and the progressive formation of the complete product make my invention practical, successful, and economical, and distinguishes it from the processes referred to above.”

Later, in speaking of the rolling of the first pour of glass, he says:

“As soon as the operation has fairly commenced, a fresh quantity of molten glass is poured between rollers, B, O, at ‘e.’ This batch of glass in its turn is rolled down upon that first poured, becoming perfectly united to it, and forming a homogeneous sheet of glass of the definite dimensions desired. Thus the operation continues progressively.”

It will be seen that the process of simultaneously pouring the first pour, rolling it, placing the netting upon it, and progressively following with and rolling the second pour were what was referred to in the reissue as “the invention common to the patents of Schmertz No. 791,216 and No. 791,217, and the patent of Leon Appert No. *81008,096.” These features were embodied in Appert’s process claim as follows:

“(1) The process of making sheets of glass with a metallic trellis imbedded therein, said process consisting in pouring and rolling out a layer of glass, simultaneously applying the trellis to the surface thereof, pouring a second layer of glass upon the trellis and rolling the same, the operation being carried on progressively, substantially as described.”

In this regard the two inventions agreed, but in the control of the wire netting they differed radically. Briefly stated, Appert initially and before any glass was poured fixed and stretched his wire in a position in which it ultimately remained after his wire-glass was made. It will thus be seen his was not a feeding operation during the glass making, and instead of the wire-web being fed to the glass the glass was, so to speak, fed to both sides of the wire. This is clearly shown in the accompanying cut, where the wire is held down at one end and at the proper horizontal plane by the hook, a, and the rule, r, and held up at the other end and at the proper horizontal plane by the spool, A, and the wheel, B, which is mounted on rails to determine the thickness of the lower sheet. In this tense, fixed, and predetermined horizontal position the wire netting has the glass fed to it from beneath by the first and from above by the second pour. So also with respect to the web in Shuman. The horizontal location of the web was rigidly fixed by the depth of the roll rib by which it was carried and imbedded in the sheet. It will thus be seen that in both Shuman and Appert there was no feeding whatever in the sense that the feed affected the possible final horizontal location of the web, but, on the contrary, in both such location was predeterminately fixed by other than feeding mechanism, and the feed was the mere source of supply. In Schmertz, however, we find the feed itself is the controlling factor in determining the ultimate horizontal location of the web, and in this feature, which was his disclosure, and in the simultaneity and progressiveness which he had in common with Appert, lie the elements *82which made his process the signal success it is. Indeed, through this protracted litigation, we have seen no reason to change the view we took of Schmertz’s process when it first came before us in Schmertz v. Appert (C. C.) 144 Fed. 117, where we said:

*81
*82“Appert and Schmertz seem to have been working at substantially the same period in developing the wire-glass process; the former in France, the latter in America. It related to casting rough plate glass for skylight and other purposes, having a wire mesh imbedded in its center. This had been done by what was known as the “European method,” which consisted of first casting the lower part of the plate, then placing the wire mesh upon it, then casting the upper part thereon. The difliculty'with that process was that the plate when finished consisted of two separate strata, the lower of which had so far congealed or solidified before the upper half was cast that the two would not unite and form a homogeneous whole. The result was they separated or split. Schmertz’s invention consisted in a process and means to apply the same; whereby the operation which rolled the lower section of the plate to the desired thinness simultaneously deposited thereon the wire mesh. Immediately thereafter the top layer was poured, and a roller followed, which reduced the sheet to the desired thickness. The speed of the operation was such that, instead of two separate plates or layers being pressed together, the two layers united to form a homogeneous whole.”

It will, of course, be noted that the originality of Schmertz with relation to his feed consisted not in its location, viz., pre-roll,’ as in patent No. 12,443, or post-roll, as in No. 791,217, for Appert’s source of wire supply was pre-roll, and Shuman’s was post-roll, but in the nature of his feed, which was one where the lower end or tip of the web depends over the lower end of support in a position to be caught between the two forming layers of glass. He for the first time disclosed, in combination with differently recessed two-roll simultaneity and progressive working, a feed which .itself and during the casting operation determined the horizontal location of the web. It was rather to seek to describe this supplying feed by a generic name than to restrict the patentee that the Patent Office referred to it as a by-gravity feed to contradistinguish it from Shuman’s and Appert’s methods of rigid wire fixation. Such being the case, it follows that the presence of this term in the claims of No. 791,217, for they are not found in the reissue of No. 791,216, is not of itself a warrant for restricting the claims of such reissue to the narrow limits of a mere improvement, but rather as a generic term which was intended to effectuate the declared purpose of the Patent Office, viz., to grant claims “covering broadly the invention common to the patents of Schmertz * * * and Appert.” The ultimate outcome of the interference between these two resulted in the grant of the Schmertz patents in suit in the proceedings hereafter noted in Schmertz v. Appert (C. C.) 144 Fed. 115. On these devices, patent No. 791,217, which showed a device for post-roll feeding, claims as follows were granted:

“(1) In the manufacture of wire-glass, the combination of a table, a leading roll, a second finishing-roll, and means for introducing the wire by gravity between said leading and finishing rolls.
“(2) In the manufacture of wire-glass, the combination of a table, a leading roll with recessed ends, a finishing-roll whose body is higher from the bed of the table than the body of the leading roll; and means for introducing the wire by gravity between said leading and finishing rolls.
*83“(3) An improvement in the process of manufacturing wire-glass, which consists in rolling a sheet of glass of less thickness than the ultimate product required, simultaneously feeding by gravity wire upon the top of said sheet at the rear of the leading roll, and rolling a second sheet ol' glass upon said original sheet and the wire, simultaneously imbedding the wire and finishing the sheet.
“(4) An improvement in the process of manufacturing wire-glass, which consists in rolling a sheet of glass of about half the ultimate thickness required, simultaneously feeding by gravity wire upon the top of said sheet at the rear of the leading roll and rolling a second sheet of glass upon said original sheet and the wire, simultaneously imbedding the wire and finishing the sheet.”

It will be noted that all of said claims have limitations confining them to post-roll feeding. And on patent No. 791,216 (reissue No. 12,443), which showed apparatus for pre-roll feed, claims (here involved) as follows were granted:

“(1) The process of making glass sheets with wire inclosed therein, consisting in simultaneously forming a layer of glass and introducing wire thereto, and completing the sheet by forming another layer upon the first layer of glass, the process being carried on progressively.
“(2) An apparatus for making sheets of glass with wire inclosed therein, consisting of a table, a leading roll to roll a layer of glass, means to support and introduce wire to the said layer, a second roll, behind the loading roll, to form a layer of glass on the first or underneath layer, the periphery of the second roll being higher above the table than that of the leading roll, and the two rolls being far enough apart to allow the glass for the second or upper layer to be poured between them.”
“(6) An improvement in the process of manufacturing wire-glass, which consists in rolling a sheet of glass of less thickness than the ultimate product required. simultaneously forcing wire upon said sheet, and forming a second sheet of glass upon said first sheet.
“(7) An improvement in the process of manufacturing wire-glass which consists in rolling a sheet of glass of about half the ultimate thickness required, simultaneously with the formation of said sheet, forcing wire in said sheet and forming a second sheet upon said first shoot.”

It will be noted that process claim 1 contains no limitation as to whether the introduction of the wire-web is pre or post roll.

Under the process of Schmertz, wire-glass has been manufactured in great quantities. Moreover, it has been found that by using the post-roll process a glass fit to grind and polish was produced. In this respect the Schmertz process brought into being--that is, in polished wire-glass — a new commercial article, viz., wire plate. Not only has a new article been produced, but the rapidity and synchronous character of Schmertz’s process rendered it possible to make much larger sheets, substantially 160 inches long and 50 inches wide. The complainant company has cast some twenty-six million feet of this glass of which a million and a half feet were afterwards polished. As an outcome of this process, polished wire-glass has also created a new article of commerce, a glass fire retardant. It was found to have remarkable capacity to resist fire, so that no matter how directly the flame impinged on it, or how intense the heat, the glass, though it cracked, did not allow flame or smoke to pass through it. Even when attacked by fire, the glass resisted intact a fire stream of 60 pounds per square inch pressure. The result has been its large use in modern high buildings where window risk is great, since it, in effect, *84serves the purpose of a shutter. Its use in elevator-cage service is now a factor in fire protection, since it prevents fire getting in the elevator shaft and making it a draft shaft to aid in the destruction of a building. The length of this opinion precludes a statement in detail of the benefits arising from the introduction of this new article which our reading of this entire record has disclosed. An examination of the prior art in the record has further satisfied us Schmertz first disclosed the process and apparatus by which the remarkable development of wire plate came about. The effective feature of Schmertz’s process is the practically synchronous or simultaneous character of his differently-recessed two-roll, two-sheet process, coupled with a successful wire feed. Molten glass responds very quickly to changes in heat conditions, and such changes render impossible results that might have been obtained an instant before. The fragile character of the wire web, the intense heat to which it is subjected, its readiness to kink, if not kept taut enough, and its readiness to stretch out of form, if subjected to tension, make the introduction of the web to the sheets a most delicate operation, and one in which the slightest de-iparture from a perfect feed results in imperfect product. Tension distorts the web, and kinking brings a loop near the surface which prevents the sheet being polished. Schmertz obtained his product by an apparatus adapted to effectuate such a synchronous process and a perfect feed, and a study of the various patents satisfies us that none of the various prior patentees disclosed in the making of wire-glass the necessity of simultaneous treatment, which in connection, be it observed, with suitable mechanism to effectuate both it and a perfect feed, made Schmertz’s process successful. These essential features in combination are lacking in the alleged anticipations.

We next come to the question of infringement. This is charged in two different machines of the respondent. In one of these we have a movable table above which are two fixed rolls so mounted that the forward one rolls a sheet of approximately one-half, and the other the whole, thickness desired. A wire web, the length of the desired sheet, is cut and held horizontally suspended between the rolls by four corner magnets. Somewhat in advance of the second roll is a cross-rod, which, with one in a corresponding position at the rear of the forward roll, keeps the wire web from coming in contact with the glass when the web is dropped by the release of the magnets. The forward end of the table being in position near the forward roll, a pour is made and the table advanced. As the table moves, the forward roll presses a sheet of one-half the thickness, which is carried forward toward the second roll. In this passage heat radiation from the sheet and contact with the air is lessened by a horizontal iron jacket which practically covers the whole sheet. As the sheet nears the second roll the magnets which hold the suspended web release and drop the web on the sheet, save where it is kept from contact therewith by the cross-rods. The second pour is then made in advance of the second roll, and on the part of the web in advance of the cross-rod, and as the table advances this stationary cross-rod lifts up the wire as it approaches, and drops or feeds it down on the going sheet in front *85of the second pour. In this way the wire web is, without tension, distortion, or kinking, drawn up and dropped, and by such feed is located both vertically and horizontally in the sheet. The jacket meanwhile conserves the heat, and a sheet is produced suitable for polishing. It is clear to us that this process embodies every element of the process claim of the reissue patent. It simultaneously pours a layer of glass and introduces the wire thereto, and it completes the sheet by forming another layer upon the first, the whole being carried on progressively. It will be observed that the claims of reissue No. 12.443 have no limitation as to whether the wire is introduced in advance of the first roll or between both rolls, but in point of fact the method of feeding over the cross-rod in respondent’s, device is a by-gravity chute feed such as is found in the claims of patent No. 791,217. The accompanying sketch, in connection with Fig. 6 of the reissue patent, shows that the fall of the web from respondent’s cross-rod as the table passes along is the same as the fall from the feed point at the foot of Schmertz’s chute. The drop from the cross-rod is a feed by gravity at that point, and that is where it is all-important. It will thus be seen the machine .embodies all the features of the first claim of No. 791,217, viz., a leading roll, a second finishing roll, and means for introducing the wire by gravity between them. We are therefore of opinion that this machine infringes claims 1, 2, 6, and 7 of the reissue, and claims 1, 2, 3, and 4 of patent No. 791,217.

The other machine is of a different type in form, but, when substance and not mere form is considered, it will be found to appropriate and reproduce the substance of Schmertz’s device. This appropriation, to our mind, consists in transposing the flat, forward movement of a flat table surface into the circular surface of a revolving roll. The difference between rolling a sheet of glass with a roller on a flat table and rolling it between two rollers is not as great as the difference in ironing a linen sheet between a hand iron and a table and ironing *86it between two rolls of a mangle. In the accompanying cut of respondent’s other device there are three rollers, A, B, and C. As we analyze the apparatus, C is the leading roll, A the second roll, and B the table roll or common base for the action of the other two rolls upon it. The wire-web comes down the chute, W, and is fed by dropping from D to V at a point between the leading roll, C, and the second roll, A. In the process the first pour, U, is made at the bite of the leading roll, C, and the table roll, B. The former is so spaced from the table roll as to roll the sheet of one-half the thickness desired. As this sheet, resting on the table roll, is carried toward the bite of the second roll, A, it meets the wire web which drops down over the tension roll, D, and the second pour, Z, being made, the lower half of the sheet is formed, and the wire is centrally imbedded by the action of the second roll, A, upon the table roll, B. The wire web coming down the chute, W, is deflected by the roll, C, and retarded by the upward movement of that roll as pressure is exerted by the tension roll, D. But from D to the glass the feed, when not unduly restrained above, is clearly one by gravity.

Now to us it is clear that, mutatis mutandis, the plane surface of a moving table being transferred to the periphery of a revolving roll, the apparatus is, in substance, transposed form, and identity of func*87tion and result, but the device of Schmcrtz. Had such a machine existed before Schmcrtz, it would have been the work, not of an inventor, but of a clever designer to have laid the periphery of the table roll on a movable table and made the first and second rolls operate on it. If we are right in finding that wire-plate is a new commercial product, and that the Schmcrtz process produced it, then the fruits of that really great invention should not be frittered away by appropriations of its substance through a transfer of .function to a mere differing but alternative mode of application. We are therefore of opinion that this device infringes the first, second, sixth, and seventh claims of the reissue and the first, second, third, and fourth claims of patent No. 791,217.

It is contended, however, that the patents in suit are invalid, that the proceedings in Appert v. Brownsville Company (C. C.) 144 Fed. 117, was not an adverse one. and therefore the decree of the court therein directing the Commissioner of Patents to issue these patents to Schinertz was void. Wc have carefully reconsidered that case, and see no reason to differ from the conclusion there reached. Before the decree was entered the fact was fully and frankly made known to the court that after the suit was brought the ownership of the Appert patent had been acquired by the adverse party. It will be observed the jurisdiction of the Circuit Court as originally invoked was undoubted. Appert filed a bill against Schmcrtz, charging infringement of his, Appert’s, patent No. 608,096. Its jurisdiction was equally clear when it entertained under Rev. St. § 491», a cross-bill by Schmcrtz, the defeated party in interference with Appert, to decree the issue of a patent in his favor. Butterworth v. Hoe. 112 U. S. 50, 5 Sup. Ct. 25, 28 L. Ed. 656; Butler v. Shaw (C. C.) 21 Fed. 321; Wheaton v. Kendall (C. C.) 85 Fed. 666; Bernardin v. Northall (C. C.) 77 Fed. 849. When, therefore, the court made its final decree, it was in a case wherein jurisdiction had been rightly acquired. It is alleged, however, that, when the two contending interests were united, the court should have dismissed the bill. We cannot so hold. The fact that there is no adverse interest of record docs not deprive the inventor of the benefit of the statute. Indeed, provision for such a contingency is made in the statute, the only requirement being that when such a proceeding is begun notice shall be given to the Commissioner of Patents. While it is true that after the merger of the two interests some additional testimony was taken, irct this testimony was only corroborative to the main, substantial proofs which were taken when Appert and Sclnnertz were most hotly contesting the interference, and which testimony had been passed upon by the Commissioner of Patents and on which he held Schmcrtz was entitled to patents. The court, therefore, having before it the testimony taken adversely, and finding the Commissioner of Patents (who under the statute was to be notified in limine had there been no adverse parties) had already considered such proofs and decided in favor of Schmcrtz, proceeded to reconsider that proof and the additional corroborative testimony, and made a decree directing the Schmcrtz patents issue. In effect, the Circuit Court simply ordered the Commissioner to do *88what he had already decided to do, and would have done had he not been prevented by the action of the Court of Appeals of the District of Columbia. We saw no reason to differ, from the decision of the Commissioner of Patents upon the facts. Nor do we now, and while, as stated, there was additional testimony taken in the Circuit Court after the adverse interests were merged, yet the correctness of that testimony is not now challenged by the respondents by themselves calling these witnesses in the present case or asking that they appear and submit to cross-examination. The Department, pursuant to its holdings as to the dates of foreign patents —De Farranti v. Westinghouse, 52 O. G. 457; Appert v. Parker, 74 O. G. 1587; American Bell Telephone Co. v. Cushman (C. C.) 57 Fed. 842; Id., 65 O. G. 135— • holding that January 12, 1894, the date of issue of Appert’s French patent, and not October 19, 1893, the date of the application, was the date Schmertz must anticipate, we did not deem it necessary in our opinion to go back of the former date, and we now feel that such date, and not October 19, 1893, is the one we should follow until the Supreme Court shall declare this departmental construction not the true one. To do otherwise would be to unsettle,a practice on which paten-tees and the public have acted. Presumably it is correct, and should not be lightly disturbed. We find no compelling reason to depart from it. In the present case, however, we have critically considered the proofs Schmertz made in the adverse interference proceedings in the way of earlier sketches whose authenticity is not questioned. These show that at their dates, in some cases prior to October 1893, Schmertz had a clear conception of the invention covered by his patents. Schmertz in the fall of 1892 had obtained copies of Shuman’s patents, and during the summer shut-down of his own plant from May 1, to July 7, 1893, took up the study of wire-glass, which was then attracting attention through Shuman’s development. This is corroborated by Schmertz’s copy of Shuman’s patent No. 483,020, bearing the stamp of its purchase in October, 1892.' On this patent are sketches of apparatus suitable for carrying out Schmertz’s process, and which Schmertz says he drew thereon during this shut-down of 1893. These sketches show on the first page a roll recessed to fix the thickness of the sheet, and on the following page a table, two rolls of different recess, a chute for introducing wire in advance of the leading roll, and two pours, one in' advance of each roll. Beneath it is written the query, “Will glass made this way weld perfectly?” This shows the device was intended to make glass, and the inquiry shows that Schmertz was keenly alive to the weak point of the European process. Schmertz also placed a sketch of his- device at that time on the Shuman patent No. 483,021, which differed from the former sketch in that he there followed Shuman in introducing his wire mesh between the rolls instead of in advance of the leading roll as in the former sketch. At the same time he made a sketch on the Brogan and Mallock patent No. 370,177, which he had also obtained the preceding October. This sketch, made by changing a figure of that patent, clearly discloses an apparatus for the practice of his process. There are a table, two rolls, and a chute marked “wire netting,” and beneath are these suggestive descriptive notes:

*89“Last roll is nearer table than second roll. A ladle of glass is cast before each roll, and wire netting fed to the glass in advance of the first roll. Wire netting will be at top of sheet, and casting will cover it up. Will it make lappy glass?”

This last inquiry is to the same effect as the prior inquiry in reference to welding, .and, like it, goes to the very point 'wherein the European theory was faulty. These conceptions of the inventions were followed by other sketches, explanations of his process to others, and actual reduction to practice within the six months following. Thus on a December stock sheet of his company Schmcrtz made a sketch between December 2d and 9th, which clearly discloses his device in the alternative form of introducing the web in advance of the leading roll and between the two. And on the January trial balance sheet of his firm he placed a sketch which showed the general form of apparatus disclosed in the preceding sketches, with the modification that in them the rolls were drawn over the glass by hand, while this last sketch showed a modification for use of machinery. It will therefore be seen that on the strength of the testimony alone, which was adversely taken in the Patent Office, the court was warranted in finding priority to Schmcrtz over Appert, and that the efficacy of that testimony as a basis of relief was not destroyed by the fact that the interests of Appert later ceased to be adverse.

But the further defense is made that the application for the patent in'suit and the invention set forth therein were abandoned by their owners, so that no valid patents could be taken. The delay in question was from the date of Schmertz’s death in October, 1898. until the fall of 1902, when his administrator began proceedings in revivor. That matter was heard by the late Judge Acheson, and evidently an order of revivor must have been made by him, although none can now be found, since in the opinion of Sclnnertz v. Appert, supra, we recited : “The cross-bill was duly revived in favor of his administrator.” We see no reason why this patent should be declared void on the ground of abandonment. The cross-bill in the Circuit Court for a decree for the issue of a patent to Schmcrtz was a case of original equity jurisdiction. It was brought within two years from the decision of the Court of Appeals of the District of Columbia, and did not involve the fatal delay of Gandy v. Marble, 122 U. S. 432, 7 Sup. Ct. 1290, 30 L. Ed. 1223. The action once begun became subject to the practice and proceedings incident to equity. It will be noted that the jurisdiction of the Circuit Court was originally invoked on Ap-pert’s bill charging infringement, and the litigants remained hostile "until March 27, 1903. From the death of Sclnnertz, the Appert interest, the complainant in the original suit, and the respondent in the cross-bill — and they were the only parties in interest, and the judgment in the case could not bind third parties. Butler v. Shaw (C. C.) 21 Fed. 321 — took no step to dismiss the cross-bill or to prosecute the suit. There was no bad faith in the delay. Moreover, there is no provision in Rev. St. § 1915 (U. S. Comp. St. 1901, p. 3392), in reference to delay in procedure, and that in Rev. St. § 4-891 (U. S. Comp. St. 1901, p. 3384), is a delay “after any action therein, of which notice shall have been given to the applicant.” Of such ac*90tion, or of such notice, there was none in this case. When, therefore, in the fall of 1902, the Schmertz interest began its proceeding to revive, Appert was in no position to charge delay to the other side. They were the first to move, and, as between the litigants, the court was certainly justified in reviving the cross-bill in favor of Schmertz’s administrator. Delay after suit is begun, which is acquiesced in by both parties, is quite different from delay by one party before suit brought. Moreover, a revivor is not a new suit, but a mere continuation of the original suit, and ordinarily a matter of right. Fitzpatrick v. Domingo, 14 Fed. 216; Hone v. Dillon (C. C.) 29 Fed. 465; Clarke v. Matthewson, 12 Pet. 164, 9 L. Ed. 1041. In addition thereto, the delay can fairly be attributed to other causes than to an intent to abandon. Schmertz left a widow and minor children. The difficulties incident to a prosecution of the claims of Schmertz were such as might well lead to delay. The widow was advised by her relatives and friends not to undertake further prosecution of the proceeding; the opposing interest was a powerful concern, and had enlisted the service of able counsel; and yet, during all that time, she certainly did no positive thing, or did her adversaries put her in a position where she was required to do anything, to evidence her intent. Moreover — -and this fact is not without significance — no rights of third parties intervened. As matters stood, Appert had prevailed in the interference, and held his patent with its claims covering Schmertz’s process, so that no one could practice it meanwhile. It has not been shown that the rights of any third parties intervened during this period df mutual inaction. We are therefore of opinion this defense cannot prevail.

In conclusion, we may say that careful consideration of this case has impressed us with the novelty and value of Schmertz’s invention. It opened up a new and broad field of activity. It brought success out of 50 years of futile effort and recognized failure. We fulfill the intent and spirit of the patent laws in affording patent protection to this, in our judgment, highly meritorious invention.

While it is not in evidence, we find confirmation of this view in a book published in 1900 by Lippincott & Co., entitled “The Wonders of Modern Mechanism,” by Charles Henry Cochrane. The author, after discussing in an article entitled “Wire Netting in Glass,” and finding English, French, and German patents nonefficient, says: “Mr. Shuman was not the first to think of embedding the wire in glass, but he was the first to concoct a method of imbedding it that was commercially practical. * * * The Shuman patents, for there were two of them, simultaneously issued under date of September 20, 1892, are the only United States patents granted for wire-glass up to that time, and constitute one of those rare instances of a process springing full-fledged and practical in a very short space of time to a most profitable and ■useful business. For the invention of a practical method of making wire-glass, Mr. Shuman .received the John Scott legacy and premium medal from the Franklin Institute in Philadelphia in 1893.”