Opinion
John S. Powers, of Buffalo, N.Y. (Fish, Richardson & Neave, Harrison F. Lyman, and Charles E. Hammett, Jr., all of Boston, Mass., of counsel), for plaintiff.
Albert R. Henry, of Buffalo, N.Y. (D. Rumsey Wheeler, of Buffalo, N.Y., of counsel), for defendant.
KNIGHT, District Judge.
This suit is for infringement of Wedger Patent No. 1,959,321, granted May 15, 1934, on an application filed December, 29, 1931. It purports to cover a so-called 'viscous softener' for activating cement applied to parts of shoe soles preparatory to cementing the parts together. Claims 1, 2, 6 and 21, product claims, and Claims 19 and 20, method claims, are in issue. Title in the plaintiff is admitted, as is the manufacture and sale of the alleged infringing compositions. Such compositions are herein known as Cataract No. 50 and Cataract No. 12. The former was brought on the market in 1934 and the latter in 1938. The practice of cementing outsoles of shoes to the uppers, as distinguished from nailing or sewing, began many years ago. In so doing pyroxylin (nitrocellulose) cement was applied along the marginal portion of each part. This was permitted to dry. It was then cut, or activated, by a solvent of cement applied by a brush, the parts assembled and held under pressure until the cement became sufficiently adhesive to hold the parts together without pressure. As early as 1915 a machine to apply evenly the cement to the two parts was devised. (Cosgrove Patent No. 1,242,925.) Rapidly increasing use of this so-called 'compo' process of shoe making brought numerous other inventions directed to facilitate production and improve methods. Ballard Patent No. 1,897,105, application filed August 17, 1929, issued February 14, 1933, shows an automatic mechanism to hold the pressure applying member in pressure applying position; Hood Patent No. 1,973,762, application filed October 1, 1929, issued September 18, 1934, shows a machine to apply the solvent simultaneously over the cement; Johnson Patent No. 1,928,693, application filed January 28, 1931, issued October 3, 1933, shows a mechanism to even the thickness of the cementing material and prevent squeeze-out, and MacKenzie Patent No. 2,014,391, application filed November 13, 1931, issued September 17, 1935, shows also a device to apply coating of cement. Rampichini Patent No. 1,048,877, application filed April 29, 1911, issued December 31, 1912, shows a method of cementing leather by the application of a solvent to cemented surfaces. In neither this patent nor in the Rampichini Patent No. 1,089,960, application filed September 6, 1912, issued March 10, 1914, is there any description of the solvent, except as 'capable of' or 'suitable for' dissolving substances. Miller Patent No. 1,757,537, application filed April 29, 1929, issued May 6, 1930, relating to a method of making shoes, simply describes the softener as a 'suitable solvent.' Ballard and Hood, supra, likewise refer simply to a solvent as an activating agent. These and various other patents were not concerned with the composition of the solvent as a means of lessening the 'dwell' in the presses.
Solvents for cement have long been and are well known. Some of such are ether and alcohol, acetone, ethyl acetate and chloroform. Nitrocellulose is a product made by treating cotton with nitric acid. Pure solvents are liquid like water. They are very volatile and on application evaporate quickly. It was known that pyroxylin when added to the solvent slowed down its activity. The greater the activity the greater the likelihood of squeeze-out and stain. The greater the viscosity of the liquid (its flow as compared with water) the greater the time the parts were required to be held in the press. The greater the time so required the larger number of presses necessary.
In or about 1928 the 'dwell' in the press was one-half hour. No less than 48 presses were required to keep one operator busy. Squeeze-out, stain, dry joint and uneven soles resulted at times. Thereafter, by the use of various expedients, the time dwell was reduced to twenty minutes. On January 12, 1931, Wedger's application for patent Serial No. 508,368 was filed. (Patent issued May 15, 1934.) This invention was directed to meet the difficulties and disadvantages hereinbefore mentioned. It was composed of nitrocellulose, together with camphor and rezyl balsam and such like materials. It was free flowing, and was called a 'blanket softener', because it formed a 'skin' on the surface to which it was applied. This skin tended to reduce the rate of evaporation, and by its use the time dwell was reduced to about five minutes. Disadvantages of squeeze-out and stain resulting from its volatility still remained. This 'blanket softener' was succeeded by the 'viscous softener' in suit, and thereby the time 'dwell' was reduced to about 70 seconds. Eight presses were needed instead of 48 as in the prior practices, squeeze-out and staining were eliminated and the amount of softening material which was required much lessened. The essence of the invention lies in the particular 'viscosity' of the softener. This is recognized in the composition of the alleged infringing softeners. Plaintiff's 'viscous softener' is extensively used by large manufacturers of shoes and has been on the market since 1932.
The defendant interposes three defenses: non-infringement; estoppel; invalidity. It also pleads a waiver and counterclaim.
The patent states the prior use of solvents to activate the cement. Among solvents most generally used are acetone, ether, and ethyl acetate. The patent recites the disadvantages in their use as softeners because of their volatility. It recites the development of patentee's 'blanket' softener and explains that the principle of the patent composition is one 'containing a relatively small quantity of a suitable cellulose derivative, preferably one which itself has a high viscosity characteristic, dissolved in a relatively volatile solvent. ' It is the nitrocellulose which gives viscosity to the composition.
Prior to 1931 the American Society for Testing Materials had adopted a method to determine the viscosity of nitrocellulose. Therefore there had been no general accepted method. This method was known and followed in the trade. It measures viscosity in seconds, reflecting the time during which a ball of a certain size falls a certain depth in a solution of nitrocellulose with a specified solvent. There were also other so-called scientific methods of measuring viscosity by different instruments, including the McMichael device or scale referred to in the patent. By means of suitable factors seconds viscosity is converted into centipoises which is a standard of measurement based on the force resistance of the composition. In the usual practice, viscosity of nitrocellulose is measured in seconds by the falling ball method and the viscosity of a solution of nitrocellulose by the instrument method.
The patent shows an example of the softener 'for use in the practice.' It comprises '40 grams on the dry basis of nitrocellulose of a nominal 1,100 seconds viscosity' combined with 900 cc of acetone and 16 cc of alcohol. It shows the viscosity of the solution according to the McMichael scale as 'about 1,000 to 1,500 centipoises.' The patent declares that the best results are obtained by use of a nitrocellulose of as high viscosity as possible with accompanying low solid content. It points out that the viscous softener or softening composition may be made by using nitrocellulose as low as about 100 seconds viscosity 'with generally satisfactory results' but that 'as the nitrocellulose viscosity goes down the concentration of nitrocellulose must be increased '. Following this the patentee points out further that the viscosity of the nitrocellulose content may be varied and how this may be done and what its effect will be. The patent points out that resulting composition shall be such that 'a ribbon of the material will remain in position when placed on dried cement without substantial flowage.'
Claim 1 is for a softener comprising a solvent 'in which is dissolved sufficiently high viscosity nitrocellulose to give the softener a plasticity such that it will stay where it is placed on the dry pyroxylin cement without substantial flowage but will flow a short distance under pressure applied thereto.'
Claim 2 is for 'A high viscosity softener * * * comprising a small amount of cellulose ester or ether dissolved in a highly volatile solvent.'
Claim 6 calls for a composition 'for cutting hardened nitrocellulose cement comprising a solvent for nitrocellulose boiling in the neighborhood of 30 to 80 degrees c and containing dissolved material including nitrocellulose of a viscosity of at least several hundred seconds, in amount such that the composition has a viscosity of above 500 centipoises.'
Claim 21 is for a composition to cut hardened cement cellulose derivative comprising a solvent of cellulose derivative boiling in the neighborhood of 30 to 80 degrees c and containing sufficient high viscosity cellulose derivative to render the cutting composition sufficiently plastic to remain where it is placed on the cement without substantial flowage when not under pressure.
Claim 19, a use claim, describes the application to dry cement of an activating agent composed of a highly viscous solution of high viscosity cellulose derivative in a low boiling solvent.
Claim 20, a use claim, is more specific than claim 19. It states that the cellulose derivative shall be as one part to 10 or 20 parts of a low boiling solvent. It states that the viscosity characteristic of the nitrocellulose shall be 'sufficiently high' to make the composition stay where put without substantial flowage.'
It is claimed that the invention, if any, lies in the specific composition shown in the example; that beyond that the scope has not been set forth with sufficient clearness (Sec. 4888, R.S., 35 U.S.C.A. § 33). The patentee follows an ordinary practice in including an example. It is essential that it be made clear that it is simply illustrative and was not intended to show the full scope of the invention. It is believed that the scope of the patent was intended to be and is broader than the example, and that this is shown by the language of the specifications. The latter point out clearly that the proportionate parts of the composition may be changed, how they may be changed, and what the resultant composition must always be. The specifications show that the patent composition may be made by using larger quantities of lower viscosity nitrocellulose provided the viscosity of the softener were kept so that it would remain in position when placed on dried cement without substantial flow. It recites that as low as 100 seconds nitrocellulose viscosity may be used. The foregoing and what has heretofore been said sufficiently point out the scope of the patent as being beyond the example. In none of the claims in suit is there found any definition of the words 'high viscosity.' There is no specific definition of 'high viscosity' in the specifications. The specifications state that 1,100 seconds viscosity is 'relatively' high viscosity. It is the contention of the plaintiff that in the trade nitrocellulose viscosity up to 5 seconds was low, 5 to 20 medium and 20 and upwards high viscosity, and therefore no specific definition or limitation in the patent was necessary to be stated. The record in this case discloses that when this patent application was filed 'high viscosity' had no definite meaning in terms of degrees in the industry; that it was more or less arbitrarily used. Prior to about 1921 there was no definite universal standard for evaluation of the viscosity characteristic. Sales were generally made by sample. There was great variance in the viscosity characteristic of nitrocellulose. With the increased use of nitrocellulose in the making of commercial lacquers, because of the varying qualities of nitrocellulose, necessity arose for the fixation of some standard. Tucker, of the Hercules Powder Company, in 1921, advocated the adoption of a method of evaluation in seconds by a falling ball method and the use of relative terms high, low and medium. He then proposed to call low viscosity from 5 to 20, medium between 40 and 60 and high between 100 and 2,400. Later, when the American Society for Testing Materials adopted a falling ball method, no relative terms of viscosity were included. In the manufacture of lacquers a viscosity less than 5 seconds was, however, known as low viscosity. In 1925, Wilson, recognized as an authority on pyroxylin and lacquers, classified pyroxylin or nitrocellulose for bronzing liquids, 30 seconds, for leather 30 to 40 seconds or up to 100 seconds and for cheaper enamels, 4 to 10 seconds. He speaks of 'high viscosity pyroxylin' as a part of a composition for bronzing. He states that the pyroxylin used in certain formulas should be 'of the high viscosity type 40 seconds or higher. ' He refers to high viscosity pyroxylin as being 'from 100 to 200 seconds' and that pyroxylin used in the manufacture of artificial leather, etc., 'is generally that of the 30 to 60 seconds variety.'
Johnson Patent No. 1,892,576, application filed April 10, 1931, speaks of high viscosity cellulose 'having viscosity substantially above 15 seconds. ' Wilson Patent No. 1,883,215, application filed July 1, 1926, speaks of high viscosity type as that having 'an apparent viscosity higher than 150 centipoises ' measured according to the method of the American Society for Testing Materials. This means 14 to 18 seconds nitrocellulose. Graham Patent No. 1,876,289, application filed September 3, 1931, describes nitrocellulose of 'relatively' high viscosity as 20-200 type. Ellis Patent No. 1,934,261, application filed August 29, 1931, refers to medium grade as something above 4 1/2 seconds. Ellis Patent No. 1,824,757, application filed December 24, 1927, calls intermediate viscosity type of being within the range of about 4 to 10 seconds viscosity. Bacon Patent No. 1,959,324, application filed September 13, 1932, speaks of high viscosity as 100 seconds or higher. The suit patent says 1,100 seconds nitrocellulose viscosity is 'relatively' high. The foregoing patents, as well as does the blanket softener patent, give some definition of the degree of viscosity as used by them. A bulletin published by the Hercules Powder Company in 1930 describes low viscosity nitrocellulose as 20 to 30 seconds and medium viscosity as 60 to 80 seconds manufacturing limits. The catalog of this company in 1931 carries the same definitions of low and medium viscosity, but it also elsewhere refers to 60-80 as quite high viscosity nitrocellulose and 30-40 seconds as high viscosity. These contradictions add to the confusion. It seems that at the present time word designations are not used by the Hercules Powder Company. It is not enough to show isolated descriptions of 30-40 seconds viscosity as high viscosity. Use in the trade means general use and that is not shown here. The patentee was required to define the meaning of 'high viscosity nitrocellulose.' Section 4888, R.S., 35 U.S.C.A. § 33. Holland Furniture Co. v. Perkins Glue Co., 277 U.S. 245, 48 S.Ct. 474, 72 L.Ed. 868; General Electric Co. v. Wabash Appliance Corp., 304 U.S. 364, 58 S.Ct. 899, 82 L.Ed. 1402. The most that can be said for it is that the specifications limit 'high viscosity nitrocellulose' as between about 100 and 1,100 seconds viscosity. It sufficiently appears that patentee so intended to limit it. He should be held to that definition.
It is argued that the patent is invalid because it is indefinite as to the amount of nitrocellulose to be used in the softener. Claim 1 describes the amount of high viscosity nitrocellulose as 'sufficient * * * to give the softener a plasticity such that it will stay where it is placed * * * . ' Claim 2 reads 'a small amount of cellulose ether * * * '. Claim 6 fixed the amount by fixing the solvent centipoises. Claim 21 describes the amount the same as in Claim 1. Claim 20 reads that cellulose derivative shall be in proportion of 1 to 10 to 20 parts of solvent, or this means 5 to 10 percent of the whole composition, and this is small as compared with the percentage of nitrocellulose in pyroxylin shoe attaching cements, which is 18 to 21 percent. In the specification is found repeated references to the amount of the nitrocellulose to accomplish a fixed result. The amount is measured by the effect of the described use. From this the patent says the solid content is something less than 5 percent and that is within the range stated in Claim 20. The specifications state that the amount may be varied and have varied. The claims also further define it by the statement as to the effect when in the softener, that is, that it must be sufficient to give a 'plasticity such that it will stay where it is placed * * * .' It is thought that the patent is sufficient in this respect. Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523, and cases there cited.
If we are right in the preceding conclusions, Claim 1 is valid only insofar as it applies to compositions containing nitrocellulose of about 100 to about 1,100 seconds, as determined by the falling ball method of the American Society for Testing Materials.
Defendant's Cataract No. 50 infringes Claim 1. Two hundred fifty nitrocellulose is 'high viscosity nitrocellulose.' Concededly the viscosity of the softener is between 1,100 and 2,300 centipoises. The solvents are all within the patent boiling point range from 30 to 80 degrees c and the percent of nitrocellulose is close to the 5 percent of the example in the patent. Defendant's Cataract No. 12 calls for 11 percent of 30 to 40 seconds nitrocellulose. While the viscosity of this softener, admittedly 1,100 to 1,900 centipoises, comes within the centipoise range of the patent, and while the solvents are low boiling and also within its range, the nitrocellulose does not come within the meaning of 'high viscosity nitrocellulose' as stated in Claim 1 as the court construes that. Defendant's Cataract No. 12 does not, therefore, infringe Claim 1.
Claim 2 presents quite a different question from Claim 1. In the first place it is broader than Claim 1. Cellulose ester or ether is substituted in the place of nitrocellulose. As described in the specifications, 'Ether and ester are each descriptive of different classes of cellulose derivatives of which nitrocellulose is one example. ' It does not require that the nitrocellulose be 'high viscosity nitrocellulose.' The specifications recite that the consistency of the softener must be such that 'the viscosity of the composition itself under existing conditions preferably should not be brought below that where a ribbon of material will remain in position when placed on dried cement without substantial flowage. ' The patent reads that the composition 'is of relatively high viscosity as compared to the solvents heretofore used, but substantially lower than the usual pyroxylin cements. ' Water has the viscosity of 1 centipoise. Sole attaching cements run from 3,400 to 4,200 centipoise. Here we find a measure of viscosity. 1,000 to 1,500 centipoise, as stated in the example, is high viscosity. This claim may be infringed by a softener composed of low, medium or high viscosity nitrocellulose in composition with the solvent specified. Both Cataract No. 50 and No. 12 have centipoise viscosity within the range stated in the example and both are viscous softeners. Both products infringe.
Claim 6 recites that the composition shall contain 'nitrocellulose of viscosity of at least several hundred seconds * * * .' This claim reads on defendant's Cataract No. 50 which shows 250 seconds viscosity nitrocellulose. It does not read on Cataract No. 12 softener. Cataract No. 50 infringes. Cataract No. 12 does not.
What has been said with reference to Claim 1 is applicable to Claims 19, 20 and 21. Cataract No. 50 infringes; Cataract No. 12 does not infringe.
There is nothing to show any use by defendant under the method claims. Defendant would not be liable as a contributory infringer. Leitch Mfg. Co. v. Barber Co., 302 U.S. 458, 58 S.Ct. 288, 82 L.Ed. 371; Carbice Corp. v. American Patents Development Corp., 283 U.S. 27, 51 S.Ct. 334, 75 L.Ed. 819; The Philad Co. v. Lechler Laboratories, Inc., 2 Cir., 107 F.2d 747.
It is asserted that the accused compositions were old and well-known in the art long prior to the filing of the patentee's application. The defendant cites many patents and certain publications. Some of these include the use of what is termed a 'solvent', many what is commonly known as cement, and others relate to nonrelated arts. Rampichini, supra; Rampichini No. 1,089,960; Miller, supra; Ballard, supra; Hood, supra, each refers to the use of a solvent to cut or activate the dry cement. None of these purport to give the composition of any solvent. No other patents are cited by the defendant as showing use of a solvent or a viscous softener as means of adhering dry cement surfaces. Ellis No. 778,232; Byers No. 1,361,961; Farrington et al. No. 1,556,512; Brick No. 1,389,575; Bernhard No. 1,895,367; Stelkens et al. No. 1,941,958; and Ellis No. 1,877,717, relate to cements used as such.
Cosgrove No. 1,242,925, MacKenzie No. 2,014,391, and Hood, Johnson and Ballard, hereinbefore mentioned, cited by the defendant as part of the prior art, relate to shoe manufacturing devices and have no relevancy here. Brick No. 1,389,575 relates to a form of an adhesive; Chalmers No. 1,112,890 to waterproof cement; Thomas et al. No. 1,505,820 to a composition to cement films; Amen No. 1,643,437 to a fabric cement; Goldman (British) No. 321,171 to a solution of cement for outer soles and uppers of shoes. These are submitted as showing the state of the art. It is not claimed that they are put to the same use as the patent nor that their composition is the same. Also, as related to the prior art, defendant cites the Hercules Powder Company pamphlet or bulletin (1931), Worden's book 'Nitrocellulose Industry' published in 1911, Wilson's 'Proxylin Enamels and Lacquers', and certain pamphlets of its own publication. The Hercules Powder Company bulletin and Wilson, supra, are concerned with the use of nitrocellulose compounds as a lacquer or finisher and Worden treats of its use in various ways. None of these three describe or refer to a softener for cement to cause surfaces to adhere. Cement, as such, has long been known, its uses and purposes. The properties of its principal constituent nitrocellulose has long been known. They have been put to uses closely related to the uses of cement. Antedating the patent it was known that some percent of nitrocellulose added to a solvent would activate cement, and that the addition of the nitrocellulose tended to lessen the fluidity of the solvent. It is urged by the defendant that collodion is a viscous softener. Undisputably collodion has been well known for many years. Collodion is claimed to be composed of 5 percent by weight of nitrocellulose and 3 parts ether and one part alcohol. Unquestionably collodion acts as an activator for cement. It is to some extent adhesive. It shows some small percent of viscosity, but it is highly volatile and flows almost like water. Its use as a softener for the purposes designed in the patent is subject to the same objections as the solvents used in various patents to which reference has been made. There is considerable other evidence in the record directed to show the prior use of nitrocellulose in solution in different degrees of concentration. The parts of the patent composition were well known. The proportion of such parts necessary to accomplish this new, novel and useful purpose was not.
The prior art discloses nothing which invalidates the patent.
The defendant claims that the plaintiff is estopped from asserting infringement because defendant cancelled and later abandoned in the Patent Office certain claims in the application for Wedger Patent No. 1,959,320 and in Wedger application Ser. No. 719,747. Defendant points attention to original Claim 17 which discloses composition for cutting nitrocellulose cement applied to shoe parts 'comprising a low boiling solvent for nitrocellulose containing a small amount of high viscosity nitrocellulose', and to Claim 19 in which nitrocellulose viscosity characteristic is stated to be 'about 300 seconds.' Admitting the rule that where there are co-pending applications by the same applicant 'all directed to common subject matter, the earlier and abandoned applications may establish estoppel against claims of the later * * * .' It is not thought that this rule is applicable here, and cases like Stearns & Co. v. Russell, 6 Cir., 85 F. 218 and Magic Light Co. v. Economy Gas-Lamp Co., 7 Cir., 97 F. 87, cited by the defendant, are not in point. Here the patents are distinct and the cancelled claims do not contain the limitations of the component parts of the two compositions. Further, the fact is that Claims 1, 2 and 6 of the viscous softener for which the application was filed December 29, 1931, was never changed in any form, and the claims of the blanket softener were not cancelled until after the patent issued on the viscous softener. Under such circumstances plaintiff is not estopped.
Defendant asserts a release and counterclaim. In October, 1935, plaintiff gave defendant notice that Cataract No. 50 infringed the patent in suit. After some correspondence the defendant wrote plaintiff stating that 'No softener cement like or substantially like that referred to in the second paragraph of your letter (no. 50) * * * will be either manufactured or sold by us. ' Defendant put its No. 12 softener on the market in 1938. It now claims plaintiff by this letter expressly released the defendant from any claim arising from the alleged infringement by selling and manufacturing Cataract No. 50 and that the letter hereinbefore referred to implies admission that No. 12 does not infringe. Defendant's reasoning makes little appeal. Obviously, however, this defense is without merit where it has been found, as here, that both Cataract No. 50 and No. 12 infringe.
This opinion may be taken as including findings of fact and conclusions of law, subject, however, to the making of any findings upon submission by either party.