Perkins Glue Co. v. Standard Furniture Co.

279 F. 458, 1922 U.S. Dist. LEXIS 878

• Court: District Court, N.D. New York
• Decided: February 4, 1922
• Status: Published

Opinion

COOPER, District Judge.

This suit is for infringements of claims 13, 28, 30, 31, and 38 of Perkins reissue letters patent No. 13,436, granted July 2, 1912, assigned to the plaintiff corporation. The patent relates to glue manufactured from carbohydrate base, cassava starch being preferred and in the practice mostly used.

[1] Before Perkins, animal glue was practically the only glue used in the art of furniture veneering, where strong joints were required. Perkins discovered that there is a point of degeneration of starch at which, when the starch so degenerated is treated to not to exceed three parts of water and a solvent there is produced wood-veneering glue as good or better than animal glue. The starch adhesives known before Perkins either contained too much water, or, if they contained the requisite small amount of water, lacked for other reasons the cohesive and adhesive strength and the spreading and flowing qualities necessary in the wood-veneering art.

Glue for wood veneering is spread uniformly by machinery, and must have proper consistency to flow through feed pipes to the spreading rolls, so as to properly penetrate the wood, as well as cover the surface. It must not penetrate too far, so as to'leave an empty joint. It must be homogeneous and nongelatinous, and, aside from its adhesiveness, it must have great cohesive strength.

The Perkins invention was very successful, as measured by the millions of pounds of glue sold each year. His prepared glue, although viscous, is amply fluid. It mixes well, and flows well through the pipes to the reservoir and rolls of the spreader, and spreads rapidly and properly over a large square area, and exhibits the peculiar degree of penetration required. Its fluidity is such as to permit of sufficient penetration to give it firm rootage in the wood pores, while its viscosity is such as to prevent excessive dissipation in penetration. Its cohesive strength is amply in excess of the requirement for a wood joint.

The Perkins patent discloses two steps: A first or degenerative step, to obtain a suitable base; and a second or final step, for converting the base into glue. The first step is not material in this case. In the second step, the dry, degenerated, or converted material- — i. e., the glue base obtained by following the first step — is mixed with two to three parts of water by weight, according to the economy and strength of the glue joint desired. Tbe batch is then agitated with a carefully regulated heat, if any. The liquid suspension thus formed is treated with a reagent, for which preferably an aqueous solution of caustic soda ‘or potash, using (with normal temperatures) from 6 to 10 per cent, of the weight of the dry powder, of dry caustic soda or equivalent of soda potash, is employed. The alkali is best added in the form of a solution of from 33 per cent, to 50 per cent, strength, to unite with one or more of the starch molecules and with the water to form a colloidal compound. The result is the Perkins glue.

Claims 28, 30, and 31 are for the ultimate prepared glue as a product. Claim 30 is a good example:

“30. A wood and fiber glue formed of a starchy carbohydrate or its equivalent by union therewith of about three parts or loss by weight of water and alkali metal hydroxide.”

Claims 13 and 38 are the process claims of making the glue. These claims are as follows:

“13. The process of making a wood glue, which consists in treating a suitable starchy product, a material portion of which is substantially insoluble in water, with a solvent of cellulose and about three parts or less by weight of water, to produce a glue having adhesive powers substantially as great as those of good animal glue.”

“38. The process of making a wood glue, which consists in suspending a glue base containing amylaceous matter insoluble in water in about three parts or less by weight of water, and treating the base with an alkali properly proportioned, substantially as described, to- produce a glue having a lasting viscosity to flow through pipes and the adhesiveness of aniinal glue.”

The defendants omit the first step, but take a cassava starch as a base, and substantially follow the second step of the Perkins patent, and get glue like Perkins glue.

The defenses are: (1) Plaintiff’s patent is not infringed; (2) plain-' tiff’s-claims are invalid, because of prior art; (3) the subject-matter was in public use more than two years before the filing of the application for a patent; (4) the claims of the reissued Perkins patent in suit are invalid, as embodying new matter inserted subsequent to the death of Perkins; (5) in view of a disclaimer filed by the plaintiff as to claims 13 and 38, these claims are invalid, or there is no infringement.

These defenses will be taken up in inverse order. As to the disclaimer (5), the Circuit Court of Appeals for the Seventh Circuit, in Solva Waterproof Glue Co. v. Perkins Glue Co., 251. Fed. 64, 163 C. C. A. 314, considered claims 13 and 38 for the process of making glue too broad, apparently because it did not appear that they exclude what was old in the art, and so held them invalid. Immediately thereafter the plaintiff corporation filed a disclaimer with the Patent Office, disclaiming from claims 13 and 38 any process of making glue, excepting where the starch or starchy carbohydrate subjected to the process is degenerated to the extent described in the patent. Thereafter Judge Sessions, in Perkins Glue Co. v. Hood et al. (D. C.) 279 Fed. 454, held claims 13 and 38, as limited by the disclaimer, to be valid and infringed. This-was followed by Perkins Glue Co. v. Holland Furniture Co. (D. C.) 279 Fed. 457, decided July 8, 1921, to the same effect. In view of the language of both the opinion of the Circuit Court of Appeals in the Solva Case and its mandate, this court feels constrained to adopt a view similar to that of Judge Sessions and hold that claims 13 and 38, as limited by the disclaimer are valid. Infringement is referred to later herein under that subject.

Upon defense (4), that the claims in suit embody new matter not within th,e scope of the original application, subscribed and sworn to by Perkins, the Circuit Court of Appeals in the Solva Case, 251 Fed. 64, 66, 163 C. C. A. 314, 316, said:

“Owing to the facts that the inventor died in 1910, before the amendments were made which enter into the claims in suit, and that such amendments were made by his executrix without their being sworn to as required by statute, appellants insist that the same are invalid, as new matter. For appellee it is insisted that said amendments are not new matter, but merely elaborations of the process and product claims of the original patents in suit. While the added matters may in a sense be new, we do not regard them of so important a character as to constitute that class of new matter which would be invalid for want of an oath under the circumstances. _ They are related to the original purposes and objects sot out in the original specification and claims, though some of them approach the border line.”

With that conclusion this court is also in accord.

Defendants’ third defense is that the Perkins patent is wholly invalid, because the subject-matter was in use more than two years before Perkins filed his application for a patent. This defense rests upon alleged public use at Poplar Bluff, Mo. The evidence as to the Poplar Bluff practices was before the 'court in the Solva Case, and has been stipulated into the record here.

In the District Court, Judge Sanborn held in the Solva Case, 223 Fed. 792, that the practice at Poplar Bluff was largely experimental, and, while the Circuit Court of Appeals did not wholly agree with this view, the court decided against the defense, and affirmed the decision on this point, on the ground that there was a substantial difference between the glue of the patent and that used at Poplar Bluff. This court feels that, in view of the substantial identity of the evidence, the decision in the Solva Case should be followed, in the absence of reasons compelling a contrary holding. Even if not controlling, defendants have failed to prove prior use at Poplar Bluff beyond a reasonable doubt.

Coming, now, to the defense numbered 2 above, the prior art, it is found that most of the patents cited by the defendants have heretofore been dismissed by the courts as not being anticipatory of the Perkins patent. He made an advancement in the art which was meritorious and far-reaching. In four decisions, three in two District Courts and one in the Circuit Court of Appeals of the Seventh Circuit it was held that the Perkins patent for glue product had not been anticipated by anything in the prior art which was cited to these courts. Of the patents in evidence here, and not in the former cases, there are but two, the French and Belgian patents, Nos. 102,200 and 34,869, respectively, of Gerard, which disclose a process and product called “Apparatine.” Gerard’s English patent for the same product was in evidence and considered in the Solva Case, but the defendants contend that the claims were not as fully stated as in the corresponding French and Belgian patents. This “Apparatine” was entirely different from the Perkins glue and wholly impracticable for wood veneering. The broad general statement in the Gerard patent that the materials may be mixed in any proportion, and a product like “joiners’ glue”' may be obtained, does not negative the patentable novelty of the Perkins product. Apparatine has been known for 30 years, and was never suitable for or used in wood veneering or similar use. It was only after Perkins had discovered the practicability of the changes in the proportion of the ingredients of the glue that his better than animal glue came into popular use. The change of proportions used by Perkins enabled him to secure a glue that was marketable and acceptable to the wood-veneering industry. These differences in kind as well as degree resulted in a product more useful than that which had theretofore been made or known. This product was not anticipated by Gerard and was patentable. Minerals Sep. Ltd. v. Hyde, 242 U. S. 261, 37 Sup. Ct. 82, 61 L. Ed. 286. Since the evidence of prior art as to the other patents has been ably discussed by Judge Sanborn, Judge Sessions, and tbe Circuit Court of Appeals, .and very little is new here, it would be needless -repetition to repeat it.

So the sole question remaining is that of infringement. Defendants’ method of making glue is as follows:

• As before stated, they do not follow the first or degenerative step at all. They take Biscamite starch glue base, obtained from the Viscamite Company, and dissolve it in about 2% parts of water, with about 3 per cent, of caustic soda, using heat to aid in solution.' Defendants also add a minute quantity of a so-called developer (a mixture of lead acetate and starch). This method, except the use of the developer, is exactly the second step of Perkins’ patent. Indeed, defendants concede that they use the second step. Moreover, defendants do not dispute that their glue thus obtained is substantially identical with Perkins glue.

[2] Whether or not the use of a developer has any effect upon the production of the glue is disputed. Even if its use has some effect, or is an improvement in the process, the defendants cannot evade infringement, if they have taken the substance of the invention of Perkins. It matters not that the addition or improvement is patented or patentable. International Time Recording Co. v. W. H. Bundy Recording Co., 159 Fed. 464, 469, 86 C. C. A. 494; Cimiotti Unhairing Co. v. American Unhairing Mach. Co., 115 Fed. 498, 504, 53 C. C. A. 230.

The defendants’ contention is that the Perkins patent in suit is limit-ed to glue made by combining the' two steps of the Perkins process, viz. (1) the degenerative step, to produce a degenerated glue base, and (2) the dissolving step, to produce the glue; that glue made in any other way is not infringement. ' They claim, further, that their glue base is not degenerated by the Perkins first step, and is of high and not of low viscosity, as is the base produced by Perkins first step; that their glue not being made from the Perkins base, or its equivalent, and therefore not by the two steps of the Perkins process, - is outside the Perkins patent, and not an infringement.

The Circuit Court of Appeals in the Solva Case did hold that the claims for the, first and second steps as such, taken separately, .were invalid. But the court, nevertheless, held the plaintiffs’ patent infringed, where, as here, a starch, coming in its commercial state, or by mechanical mixture of-commercial starches, or otherwise, degenerated to a stage equivalent to that produced by Perkins’ first step, was treated by the second step, and resulted in Perkins glue, or its equivalent.

The patentee, following - that decision, disclaimed from the second step process claims (including«claims 13 and 38) any process of making glue, excepting where the starch subjected to the process is degenerated to the extent described in the patent. To use the language of Judge Sessions in the Michigan cases (speaking of claims 13 and 38) :

“Of course the plaintiff is bound, by its disclaimer, but tbe limitation thereby imposed does not extend beyond tbe plain meaning of tbe language employed and tbe purpose sought to be accomplished. The sole object of tbe disclaimer was to narrow the claims with respect to the starchy-product or glue base to be subjected to the final process of the patent. * * * So limited, the claims are valid and infringed.”

But in the patent in suit it is claimed that raw starches, by proper adjustment of conditions, may be used without the first step treatment. Even if the patent did not so claim, doubtless the doctrine of equivalents would protect the patentee, especially in view of the change since the patent of the degree of degeneration in which the cassava starches now come on the market. Treibacher v. Chemical Co., 219 Fed. 210, 212, 135 C. C. A. 108. In the Solva Case and the Michigan cases most of the bases were not degenerated by step 1 of the Perkins patent, or by any chemical reagent, but were starches already in their commercial state, or by mixture with other raw starches, in an equivalent stage of degeneration and all those courts held the Perkins patent valid and infringed.

The defendants seek to differentiate their case here from those cases by the claim that their so-called raw starch is not degenerated to a stage equivalent to that formerly obtained by Perkins’ first step, but is, to the contrary, of the highest obtainable.viscosity, as evidenced by the fact that, when treated by the viscosimeter test, as in Perkins’ first step, it does not respond to the viscosimeter test by which, defendants assert, Perkins determined the necessary degree of degeneration to be reached by his first step. The viscosimeter test is claimed by plaintiffs to be inapplicable to a base not degenerated by the acid treatment of the first step. The plaintiffs also assert that their tests show the viscamite base to be of low viscosity.

The claim that the viscamite sold to and used by the defendants is obtained from the highest possible viscosity starch of the XXX Galohan brand largely rests on the statement of the officers and employees who sell the viscamite, and only one of these is much experienced in the starch business. It appears that these statements of high viscosity are largely opinions. True, the sample of viscamite tested at the trial did not disclose the fluidity or syrupy condition deemed desirable for glue; but plaintiff’s expert claimed to have tested the viscamite obtained from defendants’ factory, and found it did respond to tests of Perkins’ patent and produce the Perkins glue. It did not clearly appear that the sample tested in court was taken from the defendants’ premises in the presence of plaintiff (and question as to careful following of the test is also raised by the plaintiff), nor why some of the viscamite given to plaintiff’s representatives at defendants’ factory might not have been used for this test. Nor was any acetate of lead added to the mixture in this test in court, and no concurrent test with the same materials was made, using the acetate of lead developer, to show that the syrupy glue of the patent in suit could be obtained with the base used in the experiment in court.

At any rate, when the viscamite used in defendants’ factory was treated in the factory by Perkins’ second step, with about 2:l/2 parts of water, this viscamite concededly produces Perkins glue or its equivalent. This infringement, if such it be, should not be called noninfringement because of the opinion of interested witnesses as to the degree of viscosity of the starch, when that same starch, treated at the factory with Perkins’ second step, makes Perkins glue.

Plaintiff shows that substantially all cassava starches now come in a much lower state of viscosity than at the time of the Perkins patent, but professes to doubt whether this viscamite used by the defendants could be used to make this glue, unless it had had some treatment to further degenerate it. Be that as it may, the defendants’ base, however obtained, is apparently the technical and industrial equivalent of the Perkins basé. The defendants make their glue by. substantially the same process as the Perkins second step except as to the use of the developer, and evidently get the same result in glue product.

It is manifest that, where the raw starch is procured on the market, already sufficiently degenerated, so as not to require the chemical treatment of the first step of the Perkins patent, or where commercial starches of different grades and kinds may be mixed mechanically to produce the required degree of degeneration, the defendants cannot with impunity use such base, together with the other step of the Perkins process, and obtain the Perkins glue. This was in substance the holding in the Illinois and Michigan cases.

Moreover, the Perkins glue product, being the first glue produced from a vegetable base having all of the qualities of animal glue for wpod veneering, and being the first vegetable base so used, has such distinctive characteristics as to make it a clearly distinguishable product. The patent may therefore be considered from,the standpoint of its product claims, apart from the process claims therein.

[3] In a product patent, it is the duty of the court to avoid, if it reasonably can, a construction confining it to the precise method of making it which the patent happens to show. Dunn Co. v. Toronto Co., 259 Fed. 258, 261, 170 C. C. A. 326. It is very clear that the product claims are not restricted to the product made by the described process, but cover the chemical individual, however produced. Maurer v. Dickerson, 113 Fed. 870, 874, 51 C. C. A. 494; General Electric Co. v. LacoPhilips Co., 233 Fed. 96, 105, 147 C. C. A. 166. The cases cited by defendant upon this proposition, viz. Plummer v. Sargent, 120 U. S. 442, 7 Sup. Ct. 640, 30 L. Ed. 737, and Goodyear Dental Vulcanite Co. v. Davis, 102 U. S. 222, 26 L. Ed. 149, were decided upon the particular states of facts in those cases, and have little analogy to the product claims in the cáse in suit.

Where the Perkins glue, as described in the product claims (28, 30, 31), or its equivalent, is in fact obtained as described in those claims, those claims are infringed, regardless of the degree of degeneration of the-base. The question of degree of degeneration or viscosity of the base is material only to the process claims' (13 and 38) affected by the disclaimer. The Perkins patent has been infringed by the defendant corporation.

' The bill of complaint is dismissed, without costs as to the individual defendants, as there was not sufficient evidence to charge them individually with infringement.

A decree may be entered, finding-each of the claims declared upon valid and infringed, granting an injunction, directing an accounting, and referring the case to William W. Wemple, of Schenectady, N. Y., as special master, for an accounting.

Plaintiff- will recover its costs of the suit, to be taxed.