In Equity. Suit by the Remington Cash Register Company, Inc., and another, against the National Cash Register Company, to restrain infringement of patents. Decree for plaintiffs.

THOMAS, District Judge.

This is a suit brought to restrain infringement of letters patent No. 1,429,201, issued September 12, 1922, to William S. Gubelmann, for improvements in adding and recording machines. The Remington Cash Register Company, Inc., has the exclusive license thereunder to make, use, and sell cash registers. The bill charges the defendant with infringement as to a number of its various makes of cash registers, and seeks an injunction, accounting, and damages. The plaintiff withdrew its charges against all models made by defendant, except as to the so-called class 1700 and class 2000 machines, thereby materially narrowing the issues to be discussed and decided here. It is stipulated by the parties that the defendant sold in this district cash registers exemplified by certain of defendant's machines, which are in evidence, and marked Plaintiffs' Exhibits 3 and 4, respectively.

The defenses upon which the defendant relies are, broadly stated, the invalidity of the patent in suit and noninfringement. As against validity, the defendant, in its answer and brief, has set forth at length 17 reasons why the patent should be held invalid; but for the purposes of this discussion they may be grouped or summarized under four main topics or heads, viz.: First, because of undue delay in the Patent Office; second, because of abandonment; third, because of laches, estoppel, etc.; and, fourth, because of inoperativeness.

While the Gubelmann patent in suit is directed more especially to an adding machine, and defendant's machines are cash registers, adding machines and cash registers belong to the same art, and cash registers are merely adding machines adapted to specific work. Analogous — in fact, identical — mechanical organizations occur both in cash registers and in adding machines, and in the present suit adding machines and cash registers are referred to by the witnesses, without any attempt to distinguish one art from the other. The patented inventions in the Gubelmann patent which are complained of in this suit all relate to the adding and recording mechanisms of the defendant's cash register machines.

The Gubelmann Patent.

Gubelmann's patent discloses an adding machine characterized by a plurality of sets of totalizer wheels and associated printing mechanism capable of printing items introduced into the sets of totalizer wheels, and also of printing totals under control of either of the sets of totalizer wheels. It is further characterized by mechanism whereby items introduced are printed, a subtotal may be printed, and the subtotalizer cleared, without disturbing the amount standing on a grand totalizer. Another feature of the Gubelmann machine is that of a plurality of sets of totalizer wheels having the successive denominational register wheels of one set of totalizer wheels interspersed with the totalizer wheels of a totalizer of another set, so that totalizer wheels of like denominational orders are grouped together, and with the printing mechanism having elements corresponding to the denominational orders of the totalizer wheels, and arranged so that the printing mechanism may co-operate with the totalizer wheels of any set of totalizers selected by the operator. This machine also provides that, when a subtotal is printed, mechanism may be rendered effective during this cycle of operations to reset the subtotalizer wheels to zero, thus clearing the subtotalizer after the printing has occurred, and this is done without disturbing the amount standing on the grand totalizer wheels. This arrangement permits the accumulation of a grand total, the printing of an itemized record, and the printing of a subtotal thereof, without disturbing the grand total, which may be printed at the end of the day or whenever desired.

This Gubelmann machine is also so designed as to be capable of printing a grand total and in the same cycle of operations to clear the grand totalizer. The clearing operations are optional with the operator, who may print subtotals or grand totals with or without clearing these totalizers, as is desired. It was common in the art, prior to Gubelmann, to provide suitable carrying mechanism between the several denominational wheels of a totalizer, and Gubelmann might have used carrying mechanism old in the art; however, he discloses carrying mechanism of his own design, whereby, when any denominational wheel of either totalizer makes a complete revolution, the next higher order denominational wheel is advanced one step to register the necessary carry, and overflow wheels are provided and are actuated by the same kind of gear segments as are used to drive the key-controlled totalizer wheels. These outstanding characteristics may properly be designated as the law of the Gubelmann machine.

When a key is depressed in the Gubelmann machine, certain differential mechanism is rendered effective to control a pivoted adding gear segment of the denominational order corresponding to the key depressed, and to limit the movement of this gear segment to an amount corresponding to the key operated. Prior to the movement of this gear segment, it is placed in engagement with its corresponding denominational order adding wheel, and as the gear segment swings on its pivot the totalizer wheel with which the gear segment engages is turned an amount in an additive direction equal to the value of the operated key. At the same time a subtotalizer member is also turned a corresponding amount in an additive direction, whereby the amount is registered in both the grand totalizer and the subtotalizer. Printing mechanism is connected with the adding gear segments in such manner that numeral type corresponding to the differential movement of the adding gear segments are presented to the printing point, and at the proper time printing hammers are actuated to drive type against the paper on the platen and imprint a record of the item introduced into the machine during the adding operation. The gear segment is then withdrawn from the adding wheel, and in the subsequent operation of the machine the gear segment is returned to normal position. Zero stop lugs are provided to prevent the operation of any gear segment other than that which corresponds to the denominational order of the key depressed. The machine at each adding operation thus registers or adds into both a grand totalizer and subtotalizer, and also makes a printed record of the item added. These operations are repeated during a series of transactions, and totals are thus accumulated on a plurality of totalizers.

In Gubelmann's organization he provides snail cams, one for each denominational order of the totalizers. These snail cams are part of the totalizer mechanism, and rotate simultaneously with the corresponding totalizer denominational elements. When it is desired to take a total, a total key is depressed and the machine operated; but now the adding gear segments, or racks, do not come into engagement with the adding wheels. Instead, certain feeler fingers are moved forward by yielding spring connections to encounter the snail cams and set differential mechanism, which causes the proper type to be brought to the printing point to correspond to the amount standing on the totalizer with which the feeler fingers co-operate. After the type have been brought to the printing point, the printing hammers drive the type against the paper and print the total in the same manner that the item printing was accomplished. The feeler fingers are so constructed as to be capable of co-operation with either the grand totalizer or the subtotalizer, whereby the printing mechanism may be controlled to print either a grand total or a subtotal.

When it is desired to print a subtotal and preserve the grand total in the Gubelmann machine, a subtotal key is depressed. This operates mechanism to cause the feeler fingers to shift in position to co-operate with the subtotalizer cams, and if a clearing key is also depressed, certain gear mechanism rotates the shaft carrying the subtotal members, and also carrying spring pawls which pick up the subtotal members and snail cams of the subtotalizer, and turn the same to zero after the subtotal has been printed. This operation does not disturb the amount standing on the grand totalizer.

Where a grand total is to be printed by the Gubelmann machine, the grand total key is depressed, and this time the feeler fingers which control the printing type come under control of the grand totalizer denominational members, or cams, to position numeral type at the printing point corresponding to the amount standing on the grand totalizer, and the printing hammers operate to print the grand total. The operation of the machine up to this point, not only has positioned the printing type at the printing point, but has caused the gear segments to differentially descend out of engagement with the grand totalizer, so that each denominational gear segment has been stopped at a position corresponding to the amount standing in its respective denominational grand totalizer wheel. If the grand totalizer is to be cleared, the grand totalizer clearing key is now depressed. This operation brings the gear segments into mesh with the grand totalizer wheels. The gear segments move up while in engagement with the grand totalizer wheels, and each gear segment therefore rotates its corresponding grand totalizer wheel in a nonadditive direction, thereby subtracting out of the grand totalizer the amount standing thereon, so that, when the gear segments reach normal zero position, all of the denominational order wheels of the grand totalizer stand at zero, at which time the gear segments are disengaged from the grand totalizer and the grand totalizer thus stands zeroized, or cleared.

While the defendant set up hundreds of alleged prior art patents in its answer, only a few of them were referred to, and my task of considering these complicated patents has been greatly reduced by the frank admissions made by Mr. Browne, defendant's expert, who under cross-examination, on page 706 of the record, testified as follows:

"X-Q. 315. Now, Mr. Browne, will you please bear in mind the patents that you have referred to and the machines? Among the patents and machines alleged to be prior to 1900, referred to by you and introduced in evidence here, is there any single one machine or patent that discloses, first, a plurality of sets of totalizer wheels, and, second, printing mechanism for printing items introduced into the sets of totalizer wheels, and for also printing totals under control of either of said sets of totalizer wheels? A. No, sir.

"X-Q. 316. The same type of question. Among the patents and machines alleged to be prior to 1900, referred to by you and introduced in evidence here, is there any single one prior patent or machine that discloses, first, a totalizer, and, second, a subtotalizer, and, third, mechanism to accumulate items on the totalizers, and, fourth, mechanism to print said items, and to print subtotals to clear the subtotalizer, and to retain a plurality of such subtotals on the grand totalizer? A. No.

"X-Q. 317. Among the patents and machines alleged to be prior to 1900, referred to by you and introduced here in evidence, is there any single one machine or patent that discloses, first, a plurality of sets of totalizers, with the successive denominational register wheels of one set of totalizers interspersed with the register wheels of a totalizer of another set, and, second, printing mechanism having elements corresponding to the denominational order of the totalizers, and, third, mechanism for controlling the printing mechanism by the register wheels of any set of totalizers? A. No."

In view of the admitted position of the Gubelmann patent in the art, and the fact that defendant has not relied upon the prior art as being anticipatory, I do not find it necessary to analyze all of the prior art patents set forth in the record in detail, except where such prior art patents have been referred to in an attempt to limit the scope of certain claims, or where defendant has urged that it has followed the prior art constructions in its class 1700 and class 2000 machines.

Defendant's Class 1700 Machine.

Defendant's class 1700 machine embodies a subtotalizer and a grand totalizer, which are constructed so that the subtotalizer may have its wheels turned to zero without disturbing any amount standing on the grand totalizer. Printing mechanism is designed to print items introduced into the machine, and also to print subtotals of amounts standing on the subtotalizer. The subtotalizer elements comprise snail cams for each denominational order in the subtotalizer, and in printing subtotals feeler fingers are brought against the snail cams by yielding spring connections to differentially position numeral type at the printing point to correspond to amounts standing in the several denominational orders of the subtotalizer. The defendant's class 1700 machine is also constructed so that the subtotalizer members may be returned to zero or cleared, without disturbing the grand totalizer, and in the same cycle of operations in which the subtotal is printed.

The depression of the key, in defendant's class 1700 machine, causes the grand totalizer and the subtotalizer to go into engagement with adding gear segments, and then these gear segments are differentially operated, according to the value of the key depressed, to turn the members of the totalizers an amount corresponding to the key operated. After the proper amount corresponding to the depressed key has been introduced into the machine, the totalizers are disengaged from the adding gear segments, which are returned to their normal position out of engagement with the totalizers, and consequently do not disturb the amounts standing thereon. Printing type are connected, through gearing, with each denominational segmental gear, so that the movement of a segmental gear sets up, at the printing point, type numerals which correspond to the movement of the adding gear segment. Consequently, when an amount key is operated, numeral type is set up to correspond to the amount of the item introduced into the totalizers, and shortly after the proper type is presented at the printing point the printing platen drives the paper against this type, and thereby prints the amount of the item which has just been added into the machine. Since the printing type are directly gear-connected to the segmental gears, the printing types are restored to zero at the end of each adding operation, when the adding gear segments are returned to their normal position.

As previously stated, each denominational order of the subtotalizer is provided with snail cams, and when the operator depresses the subtotal key and operates the machine to take a subtotal, feeler fingers are arranged to contact with these snail cams in such manner as to position subtotal printing type at a printing point to represent subtotal amounts standing on the subtotal members. In defendant's class 1700 machine, one set of printing wheels is used to print items, and another set under control of the feeler fingers is used to print subtotals. The mechanism thrown into operation when subtotals are printed also comprises a gear member, adapted to rotate a chain of gears which communicate with a shaft carrying certain subtotalizer members, and this shaft is provided with a slot or shoulder adapted to pick up spring pawls carried by these subtotalizer denominational members, and return all of the subtotalizer denominational wheels and snail cams to zero after the subtotal has been printed, thereby clearing the subtotalizer in the same cycle of operations in which the subtotal is printed. This is all done without in any way whatever disturbing any amounts which stand upon the grand totalizer.

It will thus be seen that there is a remarkable resemblance in operation and in mechanism between the defendant's class 1700 machine and the patent in suit. While there are structural differences between them, such as, for example, the use of a single set of printing type in the Gubelmann patent to print items and subtotals, and two sets of type in defendant's 1700 machine, one for printing items and one for printing subtotals, such differences do not affect the question of infringement, provided defendant's style 1700 machine embodies inventions patented in the Gubelmann patent.

Neither the joining of two elements into an integral part, accomplishing the purpose of both, and no more, nor the separation of one integral part into two, which together do substantially what was done with the single element, will evade a charge of infringement. The rule here applicable was stated by Mr. Justice Curtis, speaking for the Supreme Court in Winans v. Denmead, 56 U.S. (15 How.) 330, at page 343, 14 L. Ed. 717, in the following manner:

"Where form and substance are inseparable, it is enough to look at the form only. Where they are separable, where the whole substance of the invention may be copied in a different form, it is the duty of courts and juries to look through the form for the substance of the invention, for that which entitled the inventor to his patent, and which the patent was designed to secure; where that is found, there is an infringement, and it is not a defense that it is embodied in a form not described, and in terms claimed by the patentee."

On the general question of infringement as to both style 1700 and style 2000 machines, the defendant calls attention to the decision of the Supreme Court in Westinghouse v. Boyden Power Brake Co., 170 U.S. 537, 18 S. Ct. 707, 42 L. Ed. 1136, where it was held that the Boyden air brake did not include in its construction the patented constructions of the Westinghouse patent. The court pointed out that, if the claims in the Westinghouse patent were to be construed so broadly as to cover the Boyden device, it would be necessary for the Westinghouse claims to then cover merely the function of the air brake and not the structure. Mr. Justice Brown, speaking for the court, on page 556 ( 18 S. Ct. 716), said:

"We find here no authority to grant a patent for a `principle' or `a mode of operation,' or an idea, or any other abstraction. A machine is a concrete thing, consisting of parts, or of certain devices and combination of devices. The principle of a machine is properly defined to be its mode of operation, or that peculiar combination of devices which distinguishes it from other machines. A machine is not a principle or an idea."

Later in the same case, at page 569 ( 18 S. Ct. 723), Mr. Justice Brown further distinguishes patentable inventions from unpatentable ideas, and on the question of infringement states the law to be that:

"Mere variations of form may be disregarded, but the substance of the invention must be there. As was said in Burr v. Duryee, 1 Wall. 531, 573 [ 17 L. Ed. 650], an infringement is a copy of the thing described in the specification of the patentee, either without variation, or with such variations as are consistent with its being in substance the same thing. If the invention of the patentee be a machine, it will be infringed by a machine which incorporates in its structure and operation the substance of the invention; that is, by an arrangement of mechanism which performs the same service or produces the same effect in the same way, or substantially the same way."

And the court concluded, as it pointed out, that the Boyden structure did not come within this rule. In the case at bar the claims of the Gubelmann patent call for structures, and not merely a principle, or functions, as claimed by defendant, and come squarely within the exceptions pointed out in the Westinghouse v. Boyden decision.

In the later case of Continental Paper Bag Co. v. Eastern Paper Bag Co., 210 U.S. 405, at page 422, 28 S. Ct. 748, 752 ( 52 L. Ed. 1122), the Supreme Court had under consideration a claim on a bag-making machine very similar in character to the Gubelmann claims here in issue, and Mr. Justice McKenna, speaking for the court, on page 422 ( 28 S. Ct. 752), said:

"The claim is not for a function, but for mechanical means to bring into working relation the folding plate and the cylinder. This relation is the very essence of the invention, and marks the advance upon the prior art. It is the thing that never had been done before, and both the lower courts found that the machines of the Continental Company were infringements of it."

The above language covers the situation in the case at bar, in that the mechanical means defined by Gubelmann's claims are the very essence of the Gubelmann invention, and defendant's machines incorporate these mechanical means, or their equivalents, in their constructions. In the Paper Bag Case, Mr. Justice McKenna, on page 418 ( 28 S. Ct. 751), further stated:

"We think it is clear that the court considered that Liddell sought to comply with section 4888 of the Revised Statutes [U.S. Comp. St. 1901, p. 3383]. In other words, he filed a description of his invention, explained its principle and the best mode in which he `contemplated applying that principle,' and did not intend to give up all other modes of application. An inventor must describe what he conceives to be the best mode, but he is not confined to that. If this were not so, most patents would be of little worth. `The principle of the invention is a unit, and invariably the modes of its embodiment in a concrete invention may be numerous and in appearance very different from each other.' Robinson on Patents, § 485."

In Eibel Process Co. v. Minnesota Ontario Paper Co., 261 U.S. 45, at page 63, 43 S. Ct. 322, 328 ( 67 L. Ed. 523), Mr. Chief Justice Taft said:

"In administering the patent law the court first looks into the art, to find what the real merit of the alleged discovery or invention is, and whether it has advanced the art substantially. If it has done so, then the court is liberal in its construction of the patent, to secure to the inventor the reward he deserves."

The Gubelmann patent, disclosing, as defendant's expert admits, broadly new inventions in the art, is entitled to a broad range of mechanical equivalents, as was held in the recent case of Hildreth v. Mastoras, 257 U.S. 27, on page 36, 42 S. Ct. 20, 24 ( 66 L. Ed. 112), where Mr. Chief Justice Taft said that, "as the Dickinson patent is a generic patent, the doctrine of broad equivalents properly applies here." This follows the well-established rule pronounced by Mr. Justice McKenna in the Paper Bag Co. Case, supra, at page 414 ( 28 S. Ct. 749), which was stated in the following language:

"The right view is expressed in Miller v. Eagle Manufacturing Company, 151 U.S. 186, 207 [ 14 S. Ct. 310, 318 ( 38 L. Ed. 121)], as follows: `The range of equivalents depends upon the extent and nature of the invention. If the invention is broad and primary in its character, the range of equivalents will be correspondingly broad, under the liberal construction which the courts give to such inventions.'"

The doctrine of equivalents is well stated by Judge Colt in Edison Electric Light Co. v. Boston Incandescent Lamp Co. (C.C.) 62 F. 397, at page 399:

"The doctrine of equivalents, as applied to primary inventions, rests upon a more satisfactory basis by the elimination of the qualification of age or time, and by holding those things to be equivalents which perform the same function in substantially the same way. The fundamental question is whether the alleged infringer makes use of the essence of the patented invention; not whether he has adopted a known equivalent, or made a patentable improvement on the invention."

The above decision was cited with approval by Judge Hough in Treibacher v. Roessler (D.C.) 214 F. 410, at 413; affirmed 219 F. 210, 135 C.C.A. 108. See, also, General Electric Co. v. Alexander (C.C.A.2d 280 F. 852; Wagner Typewriter Co. v. Wyckoff, Seamans Benedict, 151 F. 585, 81 C.C.A. 129; Manton-Gaulin Mfg. Co. v. Dairy Machinery Construction Co. (D.C.) 238 F. 210, affirmed 247 F. 317, 159 C.C.A. 411.

As against the defendant's class 1700 machine, the plaintiffs charge infringement of claims 58, 62, 213, 214, 289, 290, 292, 294, 296, and 298 of the Gubelmann patent. These claims define inventions both broadly and specifically, which may be embodied in mechanisms comprising a subtotalizer, a grand totalizer, co-operating printing mechanism, and devices co-operating therewith to accumulate items on a subtotalizer, to list these items, to print a subtotal of the items listed, and to clear the subtotalizer while accumulating totals on the grand totalizer, so that the grand totalizer registers the grand total of the groups of items which have been accumulated on the subtotalizer.

Claim 290, which is one of the more specific or narrow claims relied upon by the plaintiff, is typical of the group, and calls for:

"In a registering machine, (1) a set of subtotalizer elements; (2) a set of grand totalizer elements; (3) means for simultaneously entering items on each of said sets; (4) graduated elements movable under the control of said subtotalizer elements; (5) pivoted feeler fingers movable into engagements with said graduated elements; (6) type means adjustable under control of said feeler fingers and said graduated elements to positions corresponding to the amount standing on the subtotalizer elements; (7) means for taking an impression from said type means; (8) means for restoring said subtotalizer elements to zero position; and (9) a common operating mechanism, adapted at a single operation to impel said feeler fingers against said graduated elements, adjust said type means, take a total impression from said type means, and restore said subtotalizer elements to zero, without restoring to zero said grand total elements."

The elements of Gubelmann's claim 290 are all present in defendant's class 1700 machine, in that its construction comprises (1) a subtotalizer; (2) a grand totalizer; (3) segmental racks driven by the differential mechanism to simultaneously introduce amounts into the subtotalizer and the grand totalizer; (4) graduated snail cams forming a part of the subtotalizer mechanism and moving under control of this mechanism, so that the positions of the snail cams correspond to subtotal amounts introduced into the machine; (5) pivoted feeler fingers constructed to be yieldably brought into engagement with the snail cams when a subtotal is to be printed; (6) subtotal printing type controlled through gearing by the contact of the feeler fingers with the snail cams, so that numeral type is set up at the printing point to correspond to the amount standing on the subtotalizer; (7) a printing platen constructed to drive the paper against the subtotal numeral type, to make an impression of the subtotal standing on this type; (8) gear mechanism rendered operative during a subtotal taking operation, so that the subtotal denominational members are restored to zero after the subtotal has been printed; (9) and the machine is provided with an operating crank handle, which is a common operating mechanism that renders effective, in a single cycle of operations, the impelling of the feeler fingers against the snail cams, the adjusting of the numeral type to set up the proper amount at the printing point, the operating of the platen to cause the printing of the total, and the restoration of the subtotal denominational members to zero, all of which takes place without disturbing the amount standing on the grand totalizer.

Claim 290 is one of the claims presented in the latter part of the prosecution of the Gubelmann application, but claims to this same general subject-matter, which claims are broader in scope than claim 290, were present in the application as originally filed, and were continued in the file during the entire proceedings in the prosecution of the application; in other words, claim 290 and claims of this character are drawn to subject-matter to which Gubelmann has always asserted inventorship.

Since all of the elements of this specific claim 290 are clearly present in defendant's class 1700 machine, it is unnecessary to point out the presence of mechanism in the other nine claims which are more broadly directed to similar subject-matter which is present in defendant's class 1700 machine. See below for other nine claims.

Defendant urges, as a distinction between class 1700 machine and the Gubelmann patent, that the class 1700 machine is of the "key-operated" type, while the Gubelmann machine is of the "key-set, power-operated" type. A "key-operated" machine is one in which certain of its operations are brought about by application of power through a key lever, such as a numeral key or the like, whereas, a "key-set, power-operated" machine is one in which the operation of the key predetermines the functioning of the machine when power is applied through the operating lever, or crank, or other power-applying means; in other words, the two machines distinguish merely the manner of the application of power. In the instant case this is not material, since the matter in issue relates to combinations of mechanisms, and not to the specific manner in which power is applied to cause these mechanisms to function. Therefore, so far as the class 1700 machine is concerned, the type of the machine is a distinction without a difference.

Defendant attempts to avoid the charge of infringement, and to justify constructions of style 1700 machine, by pointing to several prior art patents whose constructions it claims to have followed in producing the 1700 machine, and directs attention to the following patents and machines; Carney, 497,860, issued May 23, 1893; Carney, 497,861, issued May 23, 1893; Carney, 748,259, issued December 29, 1903; Cleal, 718,565, issued January 13, 1903; Hiett, 580,863, issued April 20, 1897; Blessing, 602,026, issued April 5, 1898; Carney machine, Defendant's Exhibit T; and an illustrative exhibit comprising a secret total machine, Defendant's Exhibit W. The Carney patents and the Carney machine, Exhibit T, all lack any printing mechanism for printing totals, and, furthermore, these patents, and this machine, do not have a subtotalizer and a grand totalizer. The Cleal patent, 718,565, is also entirely lacking in any mechanism for printing either subtotals or grand totals. The secret total machine, Defendant's Exhibit W, was proven only as of the date of late in 1922, but testimony was offered to the effect that machines of a similar character had been sold at earlier dates.

However, defendant produced no records which satisfactorily established these earlier sales, and therefore the sales of this type of machine are not proven as of a sufficiently early date to be effective here. Line Material Co. v. Brady Electric Manufacturing Co. (D.C.) 299 F. 822. The well-settled rule, as stated by Judge Learned Hand in Kalamazoo Loose Leaf Binder Co. v. Wilson Jones L.L. Co. (D.C.) 286 F. 715, at page 717, is quite pertinent here:

"There is no documentary corroboration of it, and the testimony of the witnesses, though unimpeached, is not supported by any circumstances which put it beyond the inevitable infirmities of their recollection. The most recent declaration of the Supreme Court, in Symington v. Nat. Castings Co., 250 U.S. 383, 39 S. Ct. 542, 63 L. Ed. 1045, shows no disposition to relax the well-established canon, and I decline to consider the use as proved."

However, even if the dates of manufacture, sale, or use had been established as to this secret totalizer machine, the machine simply discloses two totalizers, one of which is open and the other of which is secret, and covered up in a locked box. These totalizers are intended to run concurrently, but, even if they should be considered as a subtotalizer and a grand totalizer, the machine lacks the very important element of printing mechanism with which to print totals.

The Hiett patent, No. 580,863, discloses stepped cams which are effective to control total printing. This patent is for an adding machine having but a single totalizer, so that it lacks the element of a grand totalizer and a subtotalizer, and, furthermore, upon cross-examination, defendant's expert, Mr. Browne, admitted that the machine disclosed in the Hiett patent is so designed that the adding wheels are not returned to zero during a total printing operation.

The Blessing patent, No. 602,026, relates to a time stamp in which there are no totalizers whatever, and the most that can be said about this patent is that it discloses merely the use of a snail cam to position a printing type. This broad concept is not claimed by the plaintiffs' patent, but the claims in suit are to combinations in an adding machine, which are not to be found in this Blessing time stamp patent.

And, finally, the defendant refers to Burroughs patent, No. 388,116, granted August 21, 1888, as having been followed in the class 1700 machine. Mr. Browne, defendant's expert, did not dwell long on this Burroughs patent. On cross-examination he admitted that the construction disclosed in this patent is not capable of printing a list of items and then printing a total of the items listed; nor is it capable of printing items and entering them on two totalizers simultaneously. Furthermore, he admitted that the printing mechanism of this Burroughs patent is not under control of either of the registers A or A ⁷, as shown and marked on the drawings of the patent.

None of these foregoing prior art machines or patents are set up as anticipations, and there is no disclosure therein which prints items, prints a subtotal of these items, and clears a subtotalizer during a subtotal printing cycle, as do both the defendant's class 1700 machine and the Gubelmann machine.

While both Gubelmann and the defendant were entitled to, and did, draw upon the mechanical arts and the prior adding machine art for old well-known mechanical elements, an analysis of the art referred to by the defendant as being that which it followed in constructing the 1700 machine shows that the very features in the class 1700 machine here complained of did not come from this old prior art, but that these features were new in the art with Gubelmann. As was admitted by the defendant's expert, defendant has not pointed to any patent, prior to the year 1900, which discloses a totalizer, a subtotalizer, mechanism to accumulate items on the totalizers, and mechanism to print said items and to print subtotals, to clear the subtotalizer, and retain a plurality of such subtotals on the grand totalizer. This is the law not only of the Gubelmann machine, but also the law of defendant's class 1700 machine.

In Sessions v. Romadka, 145 U.S. 29, at page 44, 12 S. Ct. 799, 803 ( 36 L. Ed. 609), Mr. Justice Brown, speaking for the Supreme Court in the matter of infringement of a trunk fastener patent, said:

"Both are identical in principle, operation, and design, though the trunk fasteners now in ordinary use resemble the Taylor more than the Romadka patent. In view of the fact that Taylor was a pioneer in the art of making a practical metallic trunk fastener, and invented a principle which has gone into almost universal use in this country, we think he is entitled to a liberal construction of his claim, and that the Romadka device, containing as it does all the elements of his combination, should be held an infringement, though there are superficial dissimilarities in their construction."

So here, inasmuch as the defendant's class 1700 machine embodies in its construction the patented inventions of Gubelmann, I conclude that it should be held to infringe.

Defendant's Class 2000 Machine.

Coming now to defendant's style 2000 machine, we find that the mechanical organizations to be considered are extremely intricate, and present a jungle of mechanism through which it is difficult and laborous to trace an ordered path. The work of the court, however, is reduced somewhat by the testimony of very competent experts and the assistance rendered by very able counsel, as well as by the fact that, while this machine in its entirety is complicated, it is only necessary to consider such parts of it as are in issue, and the intricacies of these mechanisms is largely due to multiplication of individual groups of devices. In each denominational order in this machine there are substantially identical groups of mechanism, and throughout the organization individual groups of totalizers, carrying devices, etc., are multiplied by the denominational capacity of the machine.

The defendant's style 2000 machine is characterized by a plurality of totalizers co-operative with a single printing mechanism which may be selectively brought into co-operation with a predetermined totalizer, so that totals may be printed as desired from any one of the many totalizers in this machine. It is further characterized by the construction in which a plurality of sets of totalizer wheels are interspersed in such manner that the denominational orders of the register wheels of one set are interspersed with the register wheels of the totalizer of another set, so that register wheels of like denominational orders are grouped together.

The machine is also adapted to print items introduced into the machine, to accumulate these items on both a subtotalizer and a grand totalizer, and to print subtotals and clear the subtotalizer in the same operation, while retaining a plurality of subtotals on a grand totalizer.

The defendant's class 2000 machine comprises a control handle and a plurality of subtotalizers and grand totalizers mounted in rows. Each time an amount is added into a subtotalizer, the same amount goes into a grand totalizer. Since a plurality of groups of totalizers are used in this machine, it is equipped with a group of keys comprising "departmental keys," which predetermine the particular totalizers into which amounts are to be entered. This is a feature of the 2000 machine which is not in issue, and a detailed description will therefore be unnecessary.

After a departmental subtotalizer and its corresponding grand totalizers have been selected by depression of the departmental key, the amount to be entered into the 2000 machine is set up on the amount keyboard by depressing the proper numeral keys, which are arranged in denominational rows like the numeral keys in the keyboard of the Gubelmann machine. This class 2000 machine is a "key-set, power-operated" type of machine, the same as the Gubelmann machine, and, after the keys have been set, the crank handle is operated to provide power for causing the parts to function.

The depression of an amount key releases the zero stop in the denominational row of the key operated to release its corresponding gear segment, but the zero stops in the denominational orders where no key is depressed remain effective to prevent movement of these gear segments, again just as in the Gubelmann machine. If the crank handle is now operated, it rotates a power shaft which causes a drive mechanism to swing the released gear segment in a counterclockwise direction until a locking member strikes the end of the numeral key which is set to stop the movement of the gear segment differentially according to the value of the amount key depressed and lock the gear segment in this position. A beam member is pivoted at one end to the gear segment wheel, and is provided at the other end with a slot which embraces a pin on an actuating link. The continued operation of the machine brings a contact member against this beam to swing the beam to its limit position. This operation causes the final position assumed by the slotted end of the beam to be dependent on the position where its corresponding gear segment is locked by the differential stop devices. The actuating link, carrying the pin which operates in the slot of the beam, is connected through suitable gearing with numeral type. The operation of the beam, the link, and gearing connections position numeral type at the printing point which correspond to the value of movement of the gear segment, and consequently represent the value of the key that has been depressed, just as the numeral type of Gubelmann's machine brought to the printing point are the same as the numeral key operated. A printing hammer platen now brings the paper into contact with the numeral type and the item introduced into the machine is printed.

After printing occurs, a grand totalizer and a subtotalizer, both determined by the departmental key which was originally selected, are brought into mesh with the gear segments, and the further continued operation of the power shaft returns the gear segments to normal position, at which time the proper denominational order wheels in the grand totalizer and in the subtotalizer are actuated by the active gear segment, thus registering the number introduced into the machine. At the end of the adding operation the grand totalizer and the subtotalizer gears are withdrawn from the gear segments, thus completing an adding operation during which the item added has been printed.

When a number of items have been introduced into the grand totalizer and subtotalizer, and each item introduced is printed, the total of these items is accumulated on both the grand totalizer and the subtotalizer. If it is desirable to print a subtotal, and clear the subtotalizer without disturbing the grand total, the control handle is moved to the proper position to select the row of the desired subtotalizer, and then the proper departmental key is depressed to select the proper subtotalizer in the row. The drive mechanism of the machine is now actuated, and the control mechanisms function to change the sequence of operations. The zero stops for all of the denominational orders are now withdrawn, and the selected subtotalizer is now thrown into mesh with the gear segments, and the actuation of these gears in a counterclockwise direction by the driving connections tends to turn the subtotalizer denominational wheels in a nonadditive direction. The zero tooth on each denominational order totalizer wheel is a long tooth, and when this tooth approaches zero, during a total-taking operation, it encounters the end of a lever which the control handle, when set for total taking, has caused to be positioned in the path of this long tooth. There is one such lever for each denominational order. This lever operates a differential stop member adapted to engage a shouldered rack associated with a corresponding gear segment wheel, and when this occurs the segmental wheel is released from its driving connection and locked in the released position. The long teeth of the subtotalizer wheels, which stand at zero, will cause these levers to lock these gear segments in zero position, whereas the subtotalizer wheels which contain amounts will permit a backward rotation on the corresponding gear segments until this rotation equals the amount which was standing on the correlated subtotalizer wheels before the total-taking operation began. When this amount has been completely subtracted by backward rotation of a subtotalizer wheel, the movement of the gear segment engaged therewith is differentially stopped by the zero tooth on the totalizer wheel striking the lever, thereby setting the differential stopping mechanism. The gear segments are now differentially stopped in such position as to exactly correspond to amounts which had been standing on the subtotalizer.

The numeral type-selecting mechanism now comes into operation, and the slotted beams carried by the gear segment wheels are operated exactly as in item printing to position numeral type at the printing point according to the position of the gear segments, or, in other words, to set up type which correspond to the amount that had been accumulated on the subtotalizer. The printing hammer platen now operates and the subtotal is printed. The further operation of the machine withdraws the subtotalizer wheels from engagement with the gear segments, and the gear segments and actuating elements are returned to normal position. It will be observed that a subtotal has been printed and the subtotalizer wheels brought to and left standing at zero, or, in other words, the subtotalizer has been cleared in the same operation that printed the subtotal; and, since the subtotalizer only was operated, this operation was performed without in any way disturbing the amount standing on the corresponding grand totalizer.

In defendant's class 2000 machine the adding gear segments are arranged around the periphery of wheels, with each wheel carrying three of these gear segments. The totalizer shafts are mounted around these wheels at 120 degrees apart, and are constructed to move toward and from the gear segment wheels, to throw the totalizers into and out of engagement with the gear segments. There are nine subtotalizers and one grand totalizer carried on each totalizer shaft, making 30 totalizers in all in this machine; that is, 27 subtotalizers (or departmental totalizers, as they may be called), and 3 grand totalizers. The totalizers on each shaft are interspersed one with another in such manner that the corresponding denominational order wheels of all the totalizers on a shaft are grouped together. When a departmental key is depressed and the crank handle is turned, it causes a shifting movement between the totalizers and the gear segments. This movement controls both printing and adding, and associates the selected totalizer with the printing mechanism in such manner as to permit a total or subtotal, as the case may be, to be printed from any one of a plurality of totalizers. Each totalizer is provided with an overflow wheel, which receives carries that are one denominational order higher than the capacity of the keyboard of the machine. These overflow wheels are actuated by gear segments exactly like those which are controlled by the keys, as in the Gubelmann machine. In view of the fact, in class 2000 machine, that a subtotalizer is effective each time a grand totalizer is adding, these overflow actuators are capable of and do co-operate with a plurality of sets of totalizers during adding operations. Furthermore, when a subtotal is printed and the machine cleared, these overflow actuators, or gear segments, zeroize or clear the overflow wheels exactly the same as similar gear segments zeroize or clear the key-controlled subtotalizer wheels. Like the Gubelmann machine, the clearing of the totalizers of defendant's class 2000 machine is optional with the operator, in that totals may be printed without clearing the totalizers as is desired.

The enlarged capacity of defendant's class 2000 machine necessarily renders the machine more complex in its organization than the machine of the Gubelmann patent. However, even in its complex form, the inventions covered by the Gubelmann claims are present, and the complexity does not relieve the class 2000 machine from being an infringement. In Machine Co. v. Murphy, 97 U.S. 120, 24 L. Ed. 935, Mr. Justice Clifford, speaking for the Supreme Court, at page 125, gave utterance to the rule which is applicable here:

"Except where form is of the essence of the invention, it has but little weight in the decision of such an issue, the correct rule being that, in determining the question of infringement, the court or jury, as the case may be, are not to judge about similarities or differences by the names of things, but are to look at the machines or their several devices or elements in the light of what they do, or what office or function they perform, and how they perform it, and to find that one thing is substantially the same as another, if it performs substantially the same function in substantially the same way to obtain the same result, always bearing in mind that devices in a patented machine are different in the sense of the patent law when they perform different functions or in a different way, or produce a substantially different result."

In speaking of the consideration to be given a primary invention, Mr. Justice Blatchford, in Morley Machine Co. v. Lancaster, 129 U.S. 263, on page 273, 9 S. Ct. 299, 302 ( 32 L. Ed. 715), said:

"Where an invention is one of a primary character, and the mechanical functions performed by the machine are, as a whole, entirely new, all subsequent machines which employ substantially the same means to accomplish the same result are infringements, although the subsequent machine may contain improvements in the separate mechanisms which go to make up the machine."

The mere difference in appearance between defendant's 2000 machine and the machine disclosed in the Gubelmann patent is no criterion as to the question of infringement, and therefore the court looks beyond the mere visual differences or similarities. Judge Coxe, speaking for the Circuit Court of Appeals, Second Circuit, in Hillard v. Fisher Book Typewriter Co. et al., 159 F. 439, at page 442, 86 C.C.A. 469, 472, observed:

"To a layman the defendants' apparatus seems very different from that employed by the patentee, but to a mechanic who looks to results rather than appearances, it probably appears quite similar. * * * In the recent case of Wagner Co. v. Wyckoff, Seamans Benedict, 151 F. 585, 592, 81 C.C.A. 129, we had occasion to examine a somewhat similar situation, where the parts of the infringing device had been so skillfully manipulated and transposed that it required the most careful analysis to discover, what we held to be the fact, that it contained every element of the patented structure and accomplished the same result."

Judge Sanborn, speaking for the Circuit Court of Appeals, Eighth Circuit, in Brammer v. Schroeder, 106 F. 918, on page 920, 46 C.C.A. 41, 43, held that:

"The general rule of the patent law is that one who invents and secures a patent for a machine or combination which first performs a useful function is thereby protected-against all machines and combinations which perform the same function by equivalent mechanical devices; but one who merely makes and secures a patent for a slight improvement on an old device or combination, which performs the same function before as after the improvement, is protected against those only who use the very device or improvement he describes and claims, or mere colorable evasions of it."

And in Lourie Implement Co. v. Lenhart et al., 130 F. 122, at page 129, 64 C.C.A. 456, 463, Judge Sanborn said:

"One may not escape infringement by adding to or subtracting from a patented device, by changing its form, or by making it more or less efficient, while he retains its principle and mode of operation, and attains its result by the use of the same or of equivalent mechanical means."

The rules above stated are quite pertinent and forceful in their application to the principles of the 2000 machine.

The Gubelmann claims in suit as to defendant's 2000 machine may conveniently be divided into seven groups, and a claim from each group selected as a typical claim.

The first group of Gubelmann claims comprises the single claim 127, which calls for: In a calculating machine, the combination of (1) a plurality of sets of accumulators; (2) means for accumulating items thereon; (3) printing devices; (4) and mechanism co-operative with said printing devices for automatically effecting printing at the will of the operator of a total from either of said sets of accumulators under control of said set of accumulators and independently of another of said sets of accumulators.

The constructions called for by Gubelmann claim 127 are present in defendant's class 2000 machine in the following combined mechanisms: (1) Any two or more of the totalizers; (2) keys of the keyboard and driving connections under control of the keys which introduce numbers into the totalizers; (3) printing devices for printing amounts entered into the machine, and also totals; (4) mechanism whereby totals are automatically printed at the will of the operator, the printing being under control of any selected one of the totalizers, which mechanism is operative to print such total without disturbing the totals standing on any other totalizer.

The second group of Gubelmann claims relates to subtotalizers and grand totalizers, and comprises claims 10, 25, 28, 50, 52, 54, 60, 61, 227, 233, 249, 250, and 287. Claim 233 is typical of this group and calls for: In a calculating machine, the combination of (1) a subtotal set of accumulators; (2) of a grand total set of accumulators; (3) and printing devices including type carriers common to said two sets of accumulators for printing the total from either set as desired. For remaining 12 claims in this group see below.

The mechanism called for by Gubelmann claim 233, was admitted by Mr. Browne, defendant's expert, to be novel with Gubelmann, and is a broad characterization of the essence of defendant's 2000 machine. This mechanism is to be found therein as follows: (1) One of the subtotalizer mechanisms; (2) one of the grand totalizer mechanisms adapted for co-operation with the subtotalizer specified; (3) the numeral printing type and the printing platen, which comprise type carriers common to either the subtotalizer or the grand totalizer to print a total from either of these totalizers at the will of the operator.

The third group of Gubelmann claims, which relate to clearing the totalizers, comprises claims 123, 223, 224, 282, 283, and 285, from which claim 283 is selected as being typical. It calls for: In a machine of the class described: (1) A plurality of accounting devices; (2) means for accumulating amounts thereon; (3) a printing mechanism; and (4) means for automatically clearing said accounting devices and for transferring their indications to said printing mechanism. For remaining five claims in this group see below.

The mechanism in defendant's class 2000 machine, which responds to Gubelmann claim 283, is as follows: (1) A plurality of totalizers; (2) the amount key control mechanism for accumulating amounts on the plurality of totalizers; (3) the printing mechanism, which prints both items and totals at the will of the operator; (4) the mechanism rendered operative by the control handle to cause a totalizer to control the printing mechanism, to transfer the amount standing thereon to the numeral type whereby this amount is printed, and which also causes the gear segments to turn the wheels of this totalizer back to zero, so that the totalizer is cleared automatically during the total printing operation.

The fourth group of Gubelmann claims, comprising mechanisms present in defendant's class 2000 machine, are claims 146, 147, 160, 247, 251, 252, 253, 254, and 257, and relate more particularly to the interspersing of one totalizer with another, whereby totalizer wheels of successive denominational orders are interspersed with the totalizer wheels of another set. Claim 253 is typical of this group and calls for: In a machine of the class described, the combination of (1) a plurality of sets of accumulators of successive denominational order, arranged so that the accumulators of one set are interspersed with the accumulators of another of said sets; (2) a printing mechanism provided with elements corresponding to the denominational orders of accumulators; and (3) means for controlling the printing elements by the accumulators of any set. For remaining eight claims of this group see below.

The mechanism in defendant's class 2000 machine, defined by claim 253, is as follows: (1) The plurality of sets of totalizers on each totalizer shaft with the successive denominational orders of one totalizer interspersed with those of the other totalizer in such manner that the totalizer wheels of like denominational orders are grouped together; (2) there are nine denominational orders in each totalizer, and there are also nine numeral type wheels of corresponding denominational value; (3) and the mechanism under control of the operator whereby these printing wheels may be brought into co-operation with any totalizer, so that the totalizer, through suitable connecting mechanism, controls the printing elements.

The fifth group of claims, comprising claims 200 and 245, relate to the overflow wheel gear segments. Claim 245 is typical of this group and calls for: In a calculating machine, the combination of (1) a plurality of sets of number keys; (2) a plurality of sets of accumulators, each set consisting of a plurality of accumulators of like denomination, there being a greater number of sets of accumulators than there are sets of keys; (3) a plurality of actuators, one for each set of accumulators, each said set of keys being correlated with for controlling a respective one of said actuators and thereby controlling one of said sets of accumulators; and (4) means for clearing any of the sets of accumulators by an operation of the machine. For the other claim in this group see below.

The mechanism in defendant's class 2000 machine, which is defined by Gubelmann claim 245, comprises (1) a plurality of sets of denominational rows of number keys; (2) the plurality of totalizers, each of which comprises a plurality of totalizer wheels, wherein each totalizer has wheels of denominational orders like those in other totalizers, and wherein each totalizer comprises nine denominational order wheels, while the keyboard contains but eight denominational sets of keys, the ninth totalizer wheel being an "overflow" wheel; (3) the nine gear segments, eight of which are correlated with the eight denominational rows of keys to permit these keys to control eight of the gear segments, the highest gear segment not being key-controlled, but co-operates with the "overflow" wheels; (4) the mechanism rendered operative by the control handle whereby, during a total taking operation, all nine gear segments are rendered effective to restore all nine totalizer wheels of a totalizer to zero during a total taking operation as to that totalizer.

The sixth group of Gubelmann claims, defining mechanism to be found in defendant's class 2000 machine, comprises claims 58, 62, 213, and 214. This group relates to mechanisms for listing items, accumulating a subtotal, printing a subtotal, and clearing a subtotalizer while retaining a plurality of subtotals on the grand totalizer. Claim 58 is typical of this group and calls for: In a calculating machine, the combination with (1) a subtotal set of registers, (2) a grand total set of registers, and (3) co-operating printing mechanism, (4) of means co-operating therewith, whereby a total of items may be accumulated on said subtotal registers, listed, its total printed, and said subtotal registers cleared at the will of the operator, and whereby a plurality of such totals may be accumulated and preserved on said grand total registers for producing a grand total of groups of items which have been accumulated on said subtotal registers. For remaining three claims of this group see below.

The mechanism of defendant's class 2000 machine which comprises the combination of this Gubelmann claim 58 is (1) the subtotalizer, comprising a set of register wheels; (2) the grand totalizer, comprising a set of register wheels; (3) the printing mechanism which co-operates with either the grand totalizer or the subtotalizer; (4) the operating mechanism comprising the finger keys, driving mechanism, differential devices, etc., whereby items are accumulated on the subtotalizer, mechanism for operating the printer to list these items, mechanism rendered operative by the control handle to print a total from the subtotalizer, and mechanism also rendered operative by the control handle to restore the wheels of the subtotalizer to zero at the will of the operator, and the constructions of this 2000 machine whereby subtotals are accumulated and preserved on the grand totalizer to represent a sum total of a plurality of subtotals.

The seventh and final group of claims of the Gubelmann patent, which designate mechanisms in defendant's class 2000 machine, are claims 55, 81, and 92, which define mechanisms for printing both subtotals and grand totals, and clearing the subtotalizer. Claim 55 is typical of these specific claims, and calls for the following combination of mechanism: In a calculating machine, the combination of (1) a subtotal accumulator, whereby a separate total of each group of a plurality of groups of numbers may be obtained; (2) a grand total accumulator whereby the grand total of a plurality of said separate totals may be obtained; (3) controlling means for determining the items to be accumulated on said subtotal and grand total accumulators; (4) actuating means for actuating both of said accumulators at a single operation of the machine to the extent determined by said controlling means; (5) means for returning said subtotal accumulator to zero independently of said grand total accumulator; (6) printing mechanism for printing the items accumulated on said subtotal and said grand total accumulators; (7) means co-operative with said printing mechanism and with said subtotal accumulator for affecting printing of a total of all the items entered in said machine since said subtotal accumulator was last cleared; and (8) means co-operative with said printing mechanism and with said grand total accumulator for effecting printing of a grand total of a plurality of said separate totals which have been accumulated on said subtotal accumulator and including all items which have been entered in said machine since said subtotal accumulator was last cleared. For the other two claims of this group see below.

The constructions in defendant's 2000 machine, which are defined in Gubelmann claim 55, are as follows: (1) A subtotalizer; (2) a grand totalizer; (3) the amount keys and connections for determining the items accumulated on the grand totalizer and on the subtotalizer; (4) the crank handle with connected mechanism, which causes the numbers set up on the keys to be introduced simultaneously into both totalizers; (5) mechanism rendered operative by the control handle to clear a subtotalizer, whereby the clearing may occur independently of any other totalizer in the machine; (6) the printing mechanism co-operative with either the subtotalizer or the grand totalizer; (7) the operating devices for the printing mechanism whereby the subtotalizer may be successively added on, its subtotals printed, and the subtotalizer cleared of numbers; (8) the construction whereby the grand totalizer retains the amounts introduced into the machine when the subtotalizer is successively added on and cleared, and whereby a grand totalizer retains the items after the last clearing of the subtotalizer, so that a grand total may be printed of all items introduced into the machine during the successive adding on and clearing of a subtotalizer, and up to the time the grand total is printed.

The defendant points out that there are similarities between certain features of the class 2000 machine and the Cook patent, No. 464,294, issued December 1, 1891, in that a positive drive for the adding racks is disclosed in said Cook patent, and that the segmental gears in class 2000 machine are also positively driven. Defendant further points to Burroughs patent, No. 504,963, granted September 12, 1893, which discloses a differential movement to the adding rack and then the engagement of the adding wheels with the rack, so that these wheels are actuated to accumulate the added number on the return movement of the racks; also in the Burroughs patent, No. 504,963, the printing type, when printing totals, is under direct control of the adding wheels, so that as these wheels are returned to zero in a nonadditive direction the movement of the actuating rack and associated type are differentially stopped to bring the proper numeral type to the printing point when the totalizer wheels are returned to zero. The Cook patent and the Burroughs patent, No. 504,963, each discloses but a single totalizer. The Cook patent has no printing mechanism. Neither of these patents are relied upon by defendant as an anticipation, but they are used merely to illustrate similarities of certain features of the class 2000 machine and the prior art, which are different from specific features of the Gubelmann machine. Such comparisons, however, or divergencies, are aside from the questions in issue, because they relate to features which are not defined by the claims in issue. These claims are for inventions which are broad in their concept, and which do not deal with the minutiæ of divergencies to which defendant directs attention.

As to the overflow wheels organization, the defendant introduces a Burroughs adding machine, and it is in evidence as Defendant's Exhibit TT. It was identified by witnesses in the employ of the Burroughs Adding Machine Company as illustrating a type of machine brought out in 1898, which is prior to the filing date of the Gubelmann patent in suit. The construction of this machine is that of a standard Burroughs adding machine having seven denominational orders in the single totalizer, and in which one denominational row of keys is omitted from the keyboard, so that, while the machine has seven denominational order wheels in the totalizer and gear segments therefor, there are but six denominational rows of keys in the keyboard. It is observed, however, that there is but a single totalizer in this machine, and that it did not teach the art how to utilize actuators for the overflow wheels of a plurality of totalizers, so that the grand totals and subtotals could be accumulated and a total printed from either of a plurality of totalizers, and wherein a totalizer is cleared as to its overflow wheels by means of actuators which comprise gear segments like those for key-controlled totalizer wheels. The defendant, in class 2000 machine, has gone far beyond the teachings of this Burroughs machine, and in doing so has come into the realm defined by the claims of the Gubelmann patent in suit.

Defendant also directs attention to the Piscicelli patent, No. 872,845, of December 3, 1907 (which was acquired by the defendant and reissued as reissue patent No. 13,632, on October 21, 1913), as disclosing the feature of totalizer selection utilized in the 2000 machine. I cannot see, however, how this patent can in any way be helpful to defendant, because it is much later than the filing date of the Gubelmann patent in suit, the application of which claimed, when filed, a grand totalizer, a subtotalizer, recording mechanism, and the shifting device controlling the recording mechanism, so that it might be shifted relative to either of the totalizers. See claim 74 of original Gubelmann file, Defendant's Exhibit 37A. It was also after the date when an application of defendant had been in interference with the application of the Gubelmann patent in suit, and therefore after the defendant had been informed of the disclosure of the Gubelmann application.

On the feature of interspersed totalizers, defendant also calls attention to the following patents: Horne, No. 357,959, issued February 15, 1887; Roberts, 487,683, issued December 6, 1892; and Fuller Griswold, No. 420,554, issued February 4, 1890. But, again, none of these patents are relied upon as being anticipations, but are referred to as indicating some particular structure similar to the defendant's class 2000 machine. An analysis of them, however, shows how fundamentally defendant has departed from any disclosure of these prior patents, and has come to those features of invention covered by the Gubelmann patent in suit. As to the Horne patent, defendant's expert, Mr. Browne, admitted, on cross-examination, that the disclosure comprised a detailed counter, and not a totalizer representing true totals, that the counter wheels could not be cleared by an operation of the machine, that the order of the arrangement of these wheels was inverse to the normal decimal system, that there is no disclosure of printing mechanism, and that the counter wheels could not be used to print a true total standing in the machine, even if a printer were disclosed. The record shows, and it is important to note, that, even, though this patent was cited by the Patent Office, nevertheless the claims of the seventh group were allowed.

The Roberts patent, while disclosing three counters, is of such construction that amounts entered into the machine go into one counter only, unless a selective key is operated, at which time amounts are entered into this counter and also into the selected counter, so that the amount standing in the first counter is not necessarily the grand total of the amounts standing on the other two counters. These counters of this Roberts patent are detail counters, and not totalizers, so that it is necessary to do mental calculating before the total can be determined. Furthermore, no provision is made for restoring the counter wheels to zero, and there is no printing mechanism disclosed therein.

The Fuller Griswold patent discloses but a single totalizer, with indicator wheels arranged between totalizer wheels. No printing mechanism is disclosed, and, since but a single totalizer is shown, obviously there is no disclosure of interspersed totalizers. The defendant also cites the Piscicelli patent, but this patent did not issue until 1907. Very clearly, then, none of these patents taught the art how to intersperse totalizers.

A similar situation was presented in Calculagraph Co. v. Wilson (C.C.) 136 F. 196, and Judge Hale, on page 197, said: "The defendant urges that his machine is merely an adaptation of the time stamp of the Emerson patent, and that it involves nothing new when compared with that patent. * * * Upon a careful examination and study of the new machine, we are satisfied that the defendant has added new elements to the Emerson patent, and has brought his device within the range of infringement. * * * The Emerson patent was not an anticipation of the patents at issue in this case."

Defendant urges that the mechanisms, in its class 1700 and class 2000 machines, are different from the mechanism disclosed in the Gubelmann patent, and that the claims of the Gubelmann patent must have read into them the specific construction disclosed by Gubelmann. It must be remembered, however, that the Gubelmann inventions, as defined by the claims and as admitted by Mr. Browne, defendant's expert, are broad in character, and that the identical mechanical organizations defined by the claims are to be found in defendant's machines.

The language of Judge Sanborn in Mast, Foos Co. v. Dempster Mill Mfg. Co., 82 F. 327, at page 333, 27 C.C.A. 191, 197, is particularly apposite to this feature of the case at bar. After quoting the claim in suit, Judge Sanborn said: "There is not an element in this combination which is not found in the windmill of the appellee, and it cannot be permitted to read other elements into this claim, and then to defeat it, because it does not use the elements it interpolates. * * * It ought not to escape here, because it does not use subordinate or unimportant elements of combinations described in other claims, which were undoubtedly omitted from this claim that the inventor might more perfectly secure the essential element of his invention."

Defendant also urges, as a distinction between its machines and the machine disclosed in the Gubelmann patent, that the subtotalizer wheels of Gubelmann are driven from the grand totalizer wheels, whereas in defendant's machine they are not so driven, and that the gear segments in the Gubelmann machine move into engagement with the totalizers, which are on stationary axes, while in defendant's machine the gear segments are on stationary axes, and the totalizers move into and out of engagement with these gear segments.

Under almost parallel conditions, Judge Lacombe, in International Time Recording Co. v. Bundy Recording Co., 159 F. 464, at page 469, 86 C.C.A. 494, 499, speaking for the Circuit Court of Appeals, Second Circuit, said: "Defendant's machine has a latch carried on the abutment, which will be depressed by putting the card squarely on the abutment, thus releasing a detent and allowing the handle which operates the printing devices to be depressed. It also has an offset on the abutment and a cut corner on the card, which secures the card's always being placed in the holder with its proper face towards the printing wheels. Both these details are lacking in the Cooper machine. Defendant has three printing wheels, instead of two. All these modifications and additions may be improvements, meritorious enough to secure a patent for them, but they will not negative infringement if a defendant uses the broad invention which a prior patentee has described and covered by his claims. Electric Smelting Co. v. Pittsburgh Reduction Co., 125 F. 933, 60 C.C.A. 636. In the Cooper patent the card receiver is movable laterally relatively to the time stamp; whereas, in defendant's machine, the card receiver is rigid and the printing wheels are movable laterally. Such a transposition of parts or of motion will not negative infringement when the same result is obtained in substantially the same way. Consolidated Fastener Co. v. Hays, 100 F. 984, 41 C.C.A. 142." See, also, Machine Co. v. Murphy, 97 U.S. 120, 24 L. Ed. 935; Morley Machine Co. v. Lancaster, 129 U.S. 263, 9 S. Ct. 299, 32 L. Ed. 715; Miehle Printing Press Mfg. Co. v. Whitlock Printing Press Mfg. Co., 223 F. 647, 139 C.C.A. 201.

But the defendant asserts that the claims in suit are void because they are for a function, and directs attention to O'Reilly et al. v. Morse et al., 15 How. 62, 14 L. Ed. 601, and other cases in support of its claim. While there are 10 claims in suit directed against the class 1700 machine, and 38 claims against the class 2000 machine, the defendant in its brief rests with a discussion of claim 213 in this connection, although it makes the broad assertion that "the other claims in issue are stricken with the same vice," without, however, showing how. Its contention is not sound. The claims were evidently drawn largely by the inventor himself, and they probably are not models of perfection in the art of drawing claims, but that does not necessarily make them invalid. They are, at least, clear, and unmistakably notify the public what the patentee claims as his own.

Claim 213, as well as the other claims in suit, are combination claims, in which the structure of certain elements is identified as "means" defined by its function. It is well established that such is a proper way to claim combinations of means. In the Paper Bag Case, 210 U.S. 405, 28 S. Ct. 748, 52 L. Ed. 1122, the Supreme Court sustained claims of a type similar to those in issue here, as has already been pointed out, and as appears on page 422 of the opinion in that case ( 28 S. Ct. 752).

The claims sustained by the Supreme Court in Morley Sewing Machine Co. v. Lancaster, supra, were of the same type as the claims here in suit; if anything, they disclose less structure than the claims under discussion. One of the claims there sustained reads:

"1. The combination, in a machine for sewing shank buttons to fabrics, of button-feeding mechanism, appliances for passing a thread through the eye of the buttons, and locking the loop to the fabric, and feeding mechanism, substantially as set forth."

The claims in suit are not for a mere function, like claim 8 in the O'Reilly v. Morse Case, supra, but set forth the very combination of means which is the contribution which Gubelmann has made in this art, and which defendant is using in its machines. I therefore conclude and hold that the patented inventions of Gubelmann are present in the defendant's class 2000 machine.

File Wrapper of Gubelmann Patent and Other Patent Office Proceedings.

The defendant asserts as a defense that the application for the Gubelmann patent was not, during the 22 years of its pendency, prosecuted diligently and with a view to securing the prompt issue of a patent thereon. The file wrapper of the Gubelmann patent in suit was introduced in evidence by the defendant and its contents are undisputed. It is impossible to read this voluminous file of 760 pages, and the other lengthy exhibits pertaining to the proceedings in the Patent Office, as I have done, without obtaining a true picture of the prodigious labors of the applicant in attempting to overcome unusual and discouraging obstacles to try to obtain protection for inventions which are admittedly new, and which inventions mark great strides in the calculating machine art.

During the prosecution of this application through the Patent Office it passed under the consideration of many different examiners, and each examiner brought forward new technical objections to the form of claims, or cited new art as to claims which had not previously been criticized or rejected, and each time a serious and conscientious effort was made, in a patient and painstaking manner, to meet the objections and rejections, only to find that in the next Office action a new and different set of difficulties was to be met. It is noteworthy, however, that throughout the entire history of this file wrapper there is not a single criticism of the applicant for lack of responsiveness to any point or rejection raised by the Patent Office. Furthermore, there never was any charge made by the Patent Office, nor is defendant raising any charge now, that any new matter was ever added to the specification or drawings of this application while it was pending. Gubelmann, time and again, attempted to completely meet the Patent Office objections and place the case in condition for allowance; but each time he was met with some further objection or rejection, which had to be overcome. From time to time lines of division were required by the Patent Office, and these had to be met.

The intricate and complicated mechanical constructions of references, which required consideration and analysis during the prosecution of this case, are of the most difficult and puzzling character, and the applicant took this work upon his own shoulders, to all of which was added the necessity of meeting the exigencies of hard-fought interferences in which the applicant prevailed. The application of the patent in suit was filed January 10, 1900, and on January 3, 1905, it was placed in interference with an application of William I. Spangler, assigned to the defendant. This interference ended by judgment for Gubelmann, who was so notified on December 11, 1905. During all the time of this Spangler interference, Gubelmann's application, which contained claims to the subject-matter here in issue, was open to defendant for inspection and to make copies. While defendant's witness, Mr. Beust, present head of the defendant's patent department, on cross-examination, asserted that no copies of the Gubelmann application could have been made after the termination of this interference, yet the testimony shows that Mr. Glass, now deceased, the predecessor of Mr. Beust, wrote a letter, dated April 28, 1913, on the letter head of the National Cash Register Company, Mr. Glass' signature having been identified by Mr. Beust, which letter refers to the file of the Gubelmann Case in the earlier interference. It is obvious that the defendant had continuously in its files a copy of the Gubelmann application from a time shortly after the declaration of the Spangler interference, and that it had been, from that time on, fully aware of the improvements in the art disclosed and claimed in Gubelmann's application.

On April 18, 1911, an interference was declared between the Gubelmann application of the patent in suit and an application of Huber C. Peters, assigned to Burroughs Adding Machine Company, and an application of Hubert Hopkins, assigned to Moon Hopkins Billing Machine Company. It appears from the testimony of defendant's witness and attorney, Mr. Beust, that over a period between the years 1899 and the year 1919 license contracts existed between the National Cash Register Company and the Burroughs Adding Machine Company, whereby the National Cash Register Company gave the Burroughs Adding Machine Company license to use their patents in adding machines, and the Burroughs Adding Machine Company gave the National Cash Register Company license to use its patents in cash registers. This Peters v. Hopkins v. Gubelmann interference, in evidence as Defendant's Exhibit 46A, a record of 807 pages, consumed the period between April 18, 1911, and February 2, 1916, or approximately five years. This interference ran its tortuous path back and forth through all the tribunals of the patent Office, and it was finally decided by the Court of Appeals in the District of Columbia in favor of Gubelmann. 44 App. D.C. 196. During the pendency of the case Gubelmann made 28 amendments to meet Patent Office requirements and 7 miscellaneous responses to Patent Office communications, totalizing in all some 450 pages, that show conclusively, as I think, a history of an earnest, sincere, and assiduous effort to secure the allowance of his application, with claims adequate to protect his novel inventions.

At one time, during the latter part of the pendency of this case, he was placed under order to show cause why the number of claims should not be reduced, and Gubelmann's answer to this order to show cause tells a story which is replete with sacrifices, of patient toil, of constant struggles, and of conscientious and sincere labor to get this case to issue. Gubelmann pointed out that he was compelled to personally take on the labor of prosecuting his own application, and that during the five years of litigation of the Peters v. Hopkins v. Gubelmann interference his opponents were two of the largest adding machine companies in the country, who employed the most experienced and skilled attorneys, who took advantage of the lack of Gubelmann's technical knowledge to drag the matter back and forth before the tribunals of the Patent Office, in an attempt to wear out Gubelmann's financial and physical resources; that he had prepared and filed divisional applications, which were required by the Examiner, and that a large number of claims in his application were claims which were taken from the application of Hubert Hopkins, and also from the application of Huber C. Peters, both of whom he had defeated in interference, which gave him the right to any claims standing in these applications which read upon his inventions as disclosed in his application; that he was also in interference with Willard L. Bundy, assigned to W.H. Bundy Recording Company, where again he was successful, and that certain of his claims came from the said Bundy application, as was entirely proper; that he was desirous of having his application passed to issue as speedily as possible, and that he would render the Patent Office every aid in his power to expedite such issue.

After this, it appears that the applicant filed amendments to try to place his case in condition for allowance, but the Patent Office still insisted that the number of claims be reduced. He responded by stating that he was desirous of reducing the number of claims copied from Hopkins and Peters, and he would do so, if he could be assured that such claims would not be allowed in the Hopkins or Peters applications, whom he had defeated in interference. In the meantime new references were again brought forward, and among them was a reference referred to in the Patent Office communication as of record, to which Gubelmann replied, and called attention to the fact that such reference was not of record, and requested further data. The Patent Office responded with the statement that the new reference was not pertinent, and therefore it was withdrawn. Gubelmann pointed out that the long pendency of the application was largely due to interference proceedings, and that he had been diligent in the prosecution of the application, even though he was taken ill in 1916, and had a very serious operation performed, which nearly cost him his life; but in spite of his illness, during which he was incapacitated and excused from prosecution, he had prepared divisional applications and made full response to the heavy actions of the Patent Office. He also called attention to the fact that it was unfortunate that the Patent Office was unable to allow Examiners sufficient time to make complete and comprehensive actions, since such incomplete actions placed a very heavy burden upon him to defend himself against references which when carefully analyzed, were found to be impertinent and irrelevant.

Finally, a complete new action was taken in the case, citing further new references, and at the same time the action was made final. Gubelmann responded to this action by an amendment, answering the citations and objections, and also appealed to the Board of Examiners in Chief. The Examiner directed the matter to the attention of the Commissioner of Patents and the appeal was withdrawn. Further rejections were made, and further amendments in answer thereto were submitted. Then it appears that in April, 1921, Gubelmann and the Examiner had finally reached an agreement. An amendment, filed April 16, 1921, states that the amendment is made in conformity with an oral interview and an agreement between the applicant and the Examiner in charge of the application, and that the application was believed to be ready for issue and an allowance solicited. Then the Patent Office wrote a letter stating that attention was called to the patent to Kettering and Chryst, No. 1,151,190, issued August 24, 1915, and requiring Gubelmann to make claim 92 of this patent within 30 days, in order that an interference might be declared.

The defendant now urges, against Gubelmann, that in June, 1921, Gubelmann made this claim from defendant's patent to Kettering and Chryst, No. 1,151,190, and that as a result an interference was declared by the Patent Office between defendant's Kettering and Chryst patent and the application of the patent in suit. Defendant emphasizes the fact that this claim was made by Gubelmann nearly six years after the grant of the Kettering and Chryst patent, and it therefore attempts to draw an inference unfavorable to Gubelmann. Defendant, however, neglected to call attention to the fact that defendant's Patent Office solicitors, Messrs. Beust Stauffer, deliberately and intentionally brought about this situation by calling attention of the Patent Office to claim 92 of their Kettering and Chryst patent and to this Gubelmann application, and asking that an interference be declared. The certified copy of the file wrapper and contents of this Kettering and Chryst patent No. 1,151,190, Defendant's Exhibit 34A, shows a paper therein contained, dated May 10, 1921, deliberately provoking this interference, although defendant already had its patent containing this claim 92. It appears that there was an interference pending between a divisional application of this Gubelmann application and this Kettering and Chryst patent, so that defendant had access to the Gubelmann patent file, and so knew that Gubelmann was making sincere and strenuous efforts to have the parent case passed to issue.

The record further shows that the defendant took exactly the same position to provoke an interference between its Osborne patent, No. 1,015,456, granted January 23, 1912, and the Gubelmann application of the patent in suit. Defendant offered in evidence a certified copy of the file wrapper and contents of this Osborne patent, Defendant's Exhibit 32A, from which it appears that a communication, dated June 6, 1921, from defendant's Patent Office solicitors, Messrs. Beust Stauffer, to the Patent Office, was omitted, but a certified copy of this communication was offered in evidence by the plaintiff, and is in evidence as Plaintiffs' Exhibit 26. This paper is similar to the one filed in the file of the Kettering and Chryst patent, and shows that the Gubelmann v. Osborne interference was also deliberately and intentionally provoked, in the face of the fact that defendant already had its patent over nine years. In the face of all this, the defendant now attempts, in a court of equity, to condemn Gubelmann because of the very situations which it deliberately brought about through its own efforts, and for the obvious purpose of keeping the application locked up, if possible, in continued interference proceedings.

Gubelmann protested against the declaration of these interferences, but was forced into them over his protest. He subsequently filed a divisional application, and put this interfering subject-matter in the divisional case, thus hoping to free the parent case for issue. But the Patent Office held up the parent case pending the outcome of these interferences. This action on the part of defendant's attorneys had the effect of delaying the issue of the patent about a year and a half.

In order to disentangle this application of the patent in suit from these self-serving interferences, provoked as they were by the defendant, Gubelmann finally chose to deliberately abandon the invention defined by the specific claims of these interferences. This brought about freeing the application of the patent in suit, so that it might be ready for issue. When the patent was about to issue, as admitted by Mr. Beust, defendant's witness and attorney, and as also appears from the decision of the Commissioner of Patents in a certified copy of the file wrapper and contents of the Gubelmann v. Osborne interference, No. 46,254, also introduced in evidence by the defendant, counsel for defendant appeared before the Commissioner of Patents, ex parte, and requested that this Gubelmann application should not be allowed to go to patent. It also appears from this decision of the Commissioner of Patents, in this exhibit of the defendant, that a petition was filed by defendant, praying that this Gubelmann application should not be allowed to issue.

The situation is one which is unusual and remarkable, to say the least. These protests and petitions appear to be based on an attack of inoperativeness against the Gubelmann construction, but, in view of Gubelmann's abandonment of the specific subject-matter in interference, these interferences terminated, and the Commissioner of Patents passed this parent application to issue, and its turbulent course through the Patent Office at last eventuated in the patent in suit.

Under all the circumstances disclosed in this record, the only reason for requesting such interferences would be that of further deliberately delaying the issuance of a patent on this Gubelmann application, and thereby further harassing Gubelmann. So long as the Gubelmann application could be held in the Patent Office, the time of its pendency would be extended until the mere weight of number of years might be used as a lever with which to crush Gubelmann's patent and destroy the results of years of prodigious labors. In the meantime, until Gubelmann's patent was granted, as defendant well knew, he could not assert any rights against any one who might be using his improvements in this complicated art.

These obstructional proceedings by the defendant did prolong the pendency of the application, and the Commissioner of Patents is to be commended for passing this application to issue as quickly as the entanglements were removed. The passing of this application to issue permitted Gubelmann to come into court for an adjudication of his rights, and where he might openly meet the defense of alleged inoperativeness, or any other defense which the defendant might raise.

In view of the fact that the defendant has continuously known of the pendency of the Gubelmann application, and of the nature and scope of the invention, since 1904 or 1905, and on account of its actions in the Patent Office looking toward delaying the Gubelmann application there, and on account of its other actions as already set forth, it is certainly in no position to claim any equities in its favor.

However, the conclusions reached are not based on those facts alone, as it appears in this record that Gubelmann, in the face of continuous opposition, did everything within his power to have the patent issue with claims commensurate with his invention. He did everything the law required him to do, and the patent would be valid, even in the absence of the defendant's knowledge and acts above set forth.

The defendant urges that the time consumed by the Gubelmann application while pending in the Patent Office worked an abandonment of his invention, if he ever had any, and it cites many authorities in support of its claim. An examination of them shows that they relate to special cases of forfeiture or estoppel, based on facts that in no way parallel the facts here. The defendant, in minute detail, points out the number of actions taken by Gubelmann, and the period of time which elapsed between each of them, but it is not seen how these references are at all pertinent or material, because under section 4894 of the Revised Statutes (Comp. St. § 9438) a definite time of one year is fixed, and Gubelmann never violated the provisions of that statute.

The Supreme Court of the United States has squarely ruled that the period while the application is pending cannot be used to weaken or limit the patent when granted. In the case of United States v. American Bell Telephone Co., 167 U.S. 224, at page 246, 17 S. Ct. 809, 812 ( 42 L. Ed. 144) Mr. Justice Brewer, speaking with reference to the rights of an inventor who abides by the provisions of the patent statutes, said:

"It will be perceived that it is conceded that some delay is unavoidable. In the very nature of things that is so. It is not possible that an application for a patent can be considered and determined on the instant. So it is not the fact, but the excessiveness of the delay of which complaint is made. The mere fact of delay does not, therefore, operate to deprive the inventor of his legal rights. Before he can be punished it must be shown that he has been guilty of a wrong — that he has caused the delay. It matters not whether the delay be reasonable or unreasonable, for a brief time or for many years, if the applicant is not responsible for it. * * * Neither can a party pursuing a strictly legal remedy be adjudged in the wrong if he acts within the time allowed, and pursues the method prescribed by the statute. If the statute gives him 5 years within which to bring an action on a note, he cannot be denied relief simply because he waits 4 years and 11 months. If he has 2 years after a judgment against him within which to take an appeal, he may wait until the last day of the 2 years.

"Under section 4886, Rev. Stat. [Comp. St. § 9430], an inventor has 2 years from the time his invention is disclosed to the public within which to make his application, and unless an abandonment is shown during that time he is entitled to a patent, and the patent runs as any other patent for 17 years from its date. He cannot be deprived of this right by proof that, if he had filed his application immediately after the invention, the patent would have been issued 2 years earlier than it was, and the public therefore would have come into possession of the free use of the invention 2 years sooner. The statute has given this right, and no consideration of public benefit can take it from him. His right exists because Congress has declared that it should. * * * A party seeking a right under the patent statutes may avail himself of all their provisions, and the courts may not deny him the benefit of a single one. These are questions not of natural but of purely statutory right. Congress, instead of fixing 17, had the power to fix 30 years as the life of a patent. No court can disregard any statutory provisions in respect to these matters on the ground that in its judgment they are unwise or prejudicial to the interests of the public. * * * Certain rules of procedure have been prescribed by the Commissioner of Patents, and a certain routine of practice has become established in that department. Now, all these matters of statutory enactment, rules of procedure and routine of practice, are things over which an applicant has no control. When he has once filed his application, complying with the statutory requirements, then the Patent Office takes possession of the matter. It determines when and how it will act, and the applicant can only ask and wait."

Again, in the Selden Patent Case, where the applicant deliberately delayed the issuance of his patent, neither the Circuit Court for the Southern District of New York nor the Circuit Court of Appeals for the Second Circuit would admit of any derogatory effect upon the Selden patent arising from the prolonged pendency of the application in the Patent Office. In sustaining the patent and finding infringement, Judge Hough had this to say about delays in the Patent Office, in Electric Vehicle Co. et al. v. C.A. Duerr Co. et al. (C.C.) 172 F. 923, at page 934:

"The claims were reworded and the specification amplified many times, and usually, after a rejection made or criticism offered by the Examiner, Selden did nothing by way of amendment or reply for about two years, the extreme limit of inactivity permitted him by the then rules of Patent Office practice. By these means he received in 1895 a patent for an invention of 1879, and in the meantime had never built a motor car, and never succeeded in getting any one sufficiently interested in his theories to experimentally try them out with larger means and better mechanical ideas than Selden himself had. * * * But the patent speaks from the date of its issue, and unless Selden did something unlawful during his 16-year wrangle with Examiners, or unless intervening American rights, available to defendants, sprang up while Selden was rewording claims, he is within the law, and his rights are the same as those of the promptest applicant.

"Without prolonging discussion, it may be held briefly that Selden did not overstep the law. He did delay. He was not in a hurry. He could not get any one to back him, and doubtless appreciated that, if he was ahead of the times, it was wise not to let his patent get ahead, too. * * * The inventor may use his discovery, or he may not; but no one else can use it for 17 years. That 17 years begins whenever the United States so decrees by its patent grant. That the applicant for patent rights acquiesces in delay, or even desires delay, is immaterial to the courts, so long as the statute law is not violated."

The Court of Appeals for the Second Circuit reversed the lower court as to the question of infringement, but on the question of the 16 years the application was pending in the Patent Office Judge Noyes, in Columbia Motor Car Co. v. C.A. Duerr Co., 184 F. 893, at page 895, 107 C.C.A. 215, 217, speaking for the Circuit Court of Appeals, said:

"It is urged that we should regard unfavorably the patent on account of this delay in the Patent Office, should seek to avoid giving it a broad construction, and should permit the alleged abuse of the law to weigh against the standing of the complainants in a court of equity. But the patentee acted wholly within his rights. He merely took advantage of the delays which the law permitted him. He followed strictly the statutes and rules of procedure, and the courts cannot exact a greater measure of diligence from him. When the patent was granted under the authority of the law, it became entitled to the consideration accorded to any other patent. If the statutes and rules permit unnecessary delays, they should be changed; but we reject the view that this court owes any duty to relieve against their operation. This patent, even if it be useful only for tribute, must be viewed without prejudice and with absolute judicial impartiality."

Mere delay does not affect the validity of the patent. Webster Electric Co. v. Podlesak (D.C.) 255 F. 907, at page 914; Electric Storage Battery Co. v. Buffalo Electric Carriage Co. (C.C.) 117 F. 314, at page 315. In Crown Cork Seal Co. v. Aluminum Stopper Co., 108 F. 845, at page 851, 48 C.C.A. 72, 78, Judge Brawley, speaking for the Circuit Court of Appeals, Fourth Circuit, in connection with subject-matter of delay, said:

"Again, it is strenuously urged that the delays in the Patent Office pending the application worked a forfeiture. The record shows that the inventor encountered serious difficulties in the securing of his patent, successive objections being interposed by the examiners, necessitating appeals. In each case, however, the necessary steps were taken within the two years limited by the statute, and during the long period which intervened between the filing of the application and the issuance of the patent, although there are abundant signs of discouragement, there are none of abandonment, for towards the end of each period of limitation he resumes the struggle, and, although often stricken down, he did by successive appeals, and in strict conformity with the requirements of the statute, at last succeed in establishing his right to the patent. * * * The inventor does not determine the measure of his rights or of his obligations. The law determines that for him, and, if the government thinks that more speed is desirable in the interest of the public, it should change the law; the courts cannot do so. Nor can they exact of inventors any degree of diligence other than compliance with the statutory provisions, official regulations, departmental requirements, and formal demands which are prescribed by and for the officials and others charged with duties under the patent laws."

Shortly after final argument counsel for defendant called the court's attention to the recent decision of the Circuit Court of Appeals for the Seventh Circuit in Overland Motor Co. v. Packard Motor Co., ___ F.2d ___, decided March 2, 1925, but that case has no real application here, because the facts are so different. In that case it appears that the patentee had taken out two earlier patents (one 13 years before the patent there in suit), which were apparently for substantially the same subject-matter, and the result would have been to prolong the monopoly. It also appears that the claims in suit were rejected repeatedly in the parent application and canceled therefrom. Later they were put into a divisional application, which became the patent. In that case, unlike the present one, the application was apparently not delayed by interferences, or by the opposition of an opponent, or by citation of new references, and the subject-matter was not of a complicated nature, like an adding machine, but related to the hub of a wheel. A reading of the case shows that the facts there are entirely different from the facts shown by this record, so the principles there enunciated have absolutely no application here.

The long pendency of the Gubelmann application cannot, therefore, under all the facts and circumstances disclosed from a complete reading and careful study of the file wrapper of the patent in suit, and of all of the many exhibits in evidence which relate to this subject-matter, be prejudicial to the patent, which must be viewed with absolute and judicial impartiality.

Abandonment.

Defendant urges that the written abandonment, as shown by Defendant's Exhibit 42A, of the claims in issue, in the interferences between Gubelmann and the Kettering and Chryst and Osborne patents, shall be expanded and extended to include substantially the entire grant of the patent in suit and the claims here in issue. The general rule of law here applicable was stated by Mr. Justice Clifford, in Jones v. Sewall, Fed. Cas. No. 7,495, 13 Fed. Cas. 1017, at page 1028, where he said: "Abandonment or dedication of an invention to the public, being in the nature of a forfeiture of a right, is not favored in law, and Mr. Justice Nelson decided that such a defense could not be sustained, unless the acts of the party invoked for the purpose were corroborated by some declarations manifesting such an intention."

In the case at bar the abandonment did not occur in the application of the patent in suit, but was made in a divisional application. This fact of itself indicates that the subject-matter abandoned was divisible; that is, separate and distinct from the subject-matter of the application of the patent in suit. However, the invention abandoned is defined by 7 claims to be found in a certified copy of the file wrapper and contents of interference No. 46,254, Gubelmann v. Osborne, and in a certified copy of the file wrapper and contents of interference No. 46,255, Kettering and Chryst v. Gubelmann, both of which certified copies were introduced by defendant.

The paper upon which defendant relies was filed as paper No. 28 in interference No. 46,255, Kettering and Chryst (which is owned by the defendant) v. Gubelmann, and is Defendant's Exhibit 54A. It reads as follows:

"In the United States Patent Office. In re Interference 46255. [Commissioner of Patents. Sept. 9, 1922.] Charles F. Kettering et al. v. William S. Gubelmann. Before the Commissioner of Patents.

"I, William S. Gubelmann, hereby declare that I abandon the invention of the following claims:

"(1) In a cash register, the combination, with a register, of a register operating mechanism, turn-to-zero devices for the register normally independent of the operating mechanism, and means for coupling the turn-to-zero devices to the operating mechanism, whereby a single regular operation of the latter will return the register to zero.

"(2) In a cash register, a series of registering disks, actuating mechanism therefor, registering amounts thereon, mechanism for resetting the disks to zero, and normally inactive connections between said actuating mechanism and said resetting mechanism.

"(3) The combination, with a register operating mechanism, of a register including a shaft and a number of registering wheels, and means for coupling the shaft to the operating mechanism to turn the wheels to their zero positions.

"(4) The combination, with a register operating mechanism, of a register, including a shaft and numbered wheels mounted thereon, and manually operated devices for coupling the shaft to the operating mechanism at will.

"(5) In a machine of the class described, the combination, with a totalizer comprising a plurality of movable elements, of operating means therefor, resetting means for the elements of the totalizer normally out of operative relation with the operating means, and means for establishing such relation, so that the elements of the totalizer may be returned to zero by actuation of the operating means.

"(6) In an accounting machine, the combination, with a plurality of totalizers, of actuating mechanism common thereto, indicators for exhibiting amounts added to any totalizer, and means for causing any totalizer to control the indicators, so as to indicate the total amount added to any totalizer.

"(7) In a calculating machine, the combination of a plurality of accumulators, a shaft, a connection between each accumulator and said shaft, an operating mechanism for said machine, actuating means operable by said operating mechanism for actuating said accumulators to accumulate numbers, and means independent of said actuating means and operable by said operating mechanism for causing said shaft to turn all of said accumulators to their initial position.

"[Signed] William S. Gubelmann."

The abandoned claims 1, 2, 3, 4, 5, and 7 define a specific resetting mechanism in a combination having but a single totalizer. The claims which Gubelmann retained, and upon which his patent was granted, are to combinations of a plurality of totalizers and other mechanisms, so that the retained claims in this patent in suit, which was issued by the Patent Office after Gubelmann's act that freed this case, so it could be issued, define inventions which are different from those identified by the claims which were abandoned. Abandoned claim 6 defines a combination including indicators for exhibiting amounts added to any totalizer and other mechanisms, all of which relate to a combination which is different from that defined by any claim in suit. The defendant attempts to draw an analogy between indicating mechanism and printing mechanism, and points to the following patents, which were introduced in evidence without any analysis or explanation by defendant's expert, to wit: Pottin, No. 312,014, issued February 10, 1885; Bates, No. 431,640, issued July 8, 1890; Clark, No. 435,400, issued September 2, 1890; Tirrell, No. 452,986, issued May 26, 1891; Pfeiffer, No. 479,720, issued July 26, 1892; and Cleal and Reinhard, No. 580,378, issued April 13, 1897. These patents to which defendant directs attention emphasize the distinction, rather than any analogy, between visual indicators and printing devices, and point out the function of each as being separate. Furthermore, in the patent in suit, the visual indicators are shown in Fig. 2 as the part numbered 238, which is quite independent and distinct from the printing mechanism of the Gubelmann machine. It therefore is clear and certain that the indicating mechanism combination of the abandoned claims 6 is entirely independent of the combinations defined by the claims here in suit, even if the abandonment had occurred in the application of the patent in suit.

It appears in the record that Gubelmann had always insisted that these abandoned claims were met by certain prior art patents, while he asserted that the claims that he retained (some of which are in suit here) were patentable. He therefore merely dropped claims which he believed were unpatentable. The rule of law applicable to this abandonment by Gubelmann is no different from the rule applicable to abandonment of any other form of personal property. In Log Owners' Booming Co. v. Hubbell, 135 Mich. 65, 97 N.W. 157, 4 L.R.A. (N.S.) 573, it was held that, where defendants claimed title to logs by abandonment, it was necessary that both the actual relinquishment and the intention to abandon be shown. Gubelmann never at any time indicated in the slightest degree either any relinquishment or intention to abandon the subject-matter of the claims of the patent here in suit. On the contrary, at the very moment he abandoned the specific inventions defined in the claims of these interferences, he was definitely, forcefully, and positively asserting inventorship to the matter here in issue. In Jones v. Sewall, supra, Mr. Justice Clifford also said: "All must agree that he did not intend to dedicate it to the public, as his application for a patent was then pending in the Patent Office, and the evidence shows that he continued to press it, with confident hopes of success, until the adverse decision was announced."

The formal abandonment filed by Gubelmann in the above-mentioned interferences, Defendant's Exhibits 53A and 54A, show on their face, that the abandonment is specific. It should be noted that Exhibit 54A specifically states that "I abandon the invention of the following claims," and then follows a full recital of the specific claims that are abandoned. In Reed v. Cropp Concrete Machinery Co. et al., 239 F. 869, 152 C.C.A. 653, where an applicant avoided an interference by disclaimer, Judge Alschuler held that such a disclaimer only extends to the particular claim or claims as to which the interference had been declared, and the judgment following does not affect the question of invention as to other claims, although they may contain other interfering matter.

The abandonment by Gubelmann is definitely limited to the specific inventions defined by the claims recited in his formal abandonment, and does not go beyond the limits of those specific claims, and therefore has no effect on the claims of the patent in suit which are to a different subject-matter.

Laches, Estoppel, etc.

After having had an interference with the application of the patent in suit as early as 1905, and after having made strenuous efforts to prevent the issuance of the patent, defendant now urges that the long pendency of this application should deprive Gubelmann of the fruits of his inventions and of the years of toil, and of his victories in interferences with defendant and its licensor licensee, the Burroughs Adding Machine Company. The application pended in the Patent Office 22 years through no fault of Gubelmann, who was compelled to meet the changing positions of Examiners, the analyses of new references, and the unusually intricate and technical problems raised by his opponents in the several interferences in which this application was from time to time involved. Defendant now directs attention to Gubelmann claims 55, 58, 60, 127, 213, 224, 233, 283, 290, 292, and 296, which it asserts are barred to Gubelmann for late presentation in his application, and because of the marketing by the Burroughs Adding Machine Company of a machine known as the Burroughs duplex machine, or because of a defendant's machine called the Union Dime Savings Bank machine, or because of defendant's class 1700 and class 2000 machines.

In the first place, the subject-matter of the several claims in the Gubelmann patent which were above mentioned relate to subject-matter to which Gubelmann has always asserted inventorship, both at and since filing his application, which was years prior to these alleged intervening structures. At different times, in somewhat different language, but always the same in substance, and with the same insistence, Gubelmann persisted, throughout the pendency of his application, that it was the substance of the invention claimed, and not the mere form of the claim, which should be scrutinized.

In the second place, there can be no laches or estoppel, or intervening rights, against one who is prior in his assertion of inventorship to the subject-matter in question, and who has continuity in his claims before, during, and after the dates when the alleged intervening rights were supposed to be incubating, for, as Mr. Justice Clifford said in Jones v. Sewall, supra, at page 1028: "Delays in the Patent Office, which an inventor cannot prevent, will not impair his title to his invention; nor can any use of the invention during such delays, if without his consent and allowance, afford any evidence to support the issue that the inventor abandoned the invention to the public."

It is stipulated by counsel that, as to the Burroughs duplex machine, the first one was sold in the spring of 1909. This is long after Gubelmann had asserted inventorship to the subject-matter of the claims here under consideration, and is after the defendant had access to the Gubelmann application of the patent in suit through the Spangler interference in 1905. From the stipulation it also appears that the Burroughs Adding Machine Company has continued to manufacture and sell machines known as the "Burroughs duplex" down to the present time, even though Huber C. Peters, whose application was owned by the Burroughs Adding Machine Company, was defeated by Gubelmann in the Hopkins v. Peters v. Gubelmann interference, which terminated in 1916 by a decision of the Court of Appeals of the District of Columbia, sustaining the Commissioner of Patents, who in turn sustained the Board of Examiners in Chief in favor of Gubelmann.

As suggested by plaintiffs, this Burroughs duplex machine appears to be an embodiment of the very invention which, after a vigorous contest, was awarded to Gubelmann, and not to Peters. It is inconceivable how it can be successfully contended, for or on behalf of the Burroughs Adding Machine Company, that it acquired any "intervening rights." Furthermore, as previously stated, the testimony of defendant's witness, Mr. Beust, on cross-examination, shows that the Burroughs Adding Machine Company and defendant had reciprocal patent licenses from 1899 to 1919.

Defendant's Exhibit 83A, the Union Dime Savings Bank machine, was shipped in August, 1910. The testimony shows that the first use of this machine was an experiment, and that the defendant regarded it as such, because Mr. J.H. Patterson, the then president of the defendant company, said to the secretary of the bank as late as December, 1910, that whatever was paid for the machine "will be a contribution to our invention department." It also appears that the machine gave some trouble, and required replacements and frequent repairs, at least up to the year 1913. Furthermore, it appears that the mechanism was entirely inclosed in a metal casing, and that the machine was never accessible to the public, even for operation, and certainly not for inspection. Any date that can be assigned to this machine is years after the time of the Spangler interference in 1905, when defendant was fully acquainted with the Gubelmann application of the patent in suit, and the real commercial use began after the date of the Hopkins v. Peters v. Gubelmann interference with defendant's licensee licensor. The existence of this machine at the dates assigned to it is of no avail to defendant.

Defendant's witness Beust testified that defendant's class 1700 machine was first put out in December, 1919, and that defendant's class 2000 machine was put out in April, May, or June of 1921. It is noteworthy to observe that it was about this time the defendant was particularly active in attempting to block the issuance of the patent in suit. The dates of both the class 1700 and class 2000 machines are not only many years after the Spangler interference in 1905, and the Hopkins v. Peters v. Gubelmann interference in 1911, but it is also after other interferences had been declared with divisions of the Gubelmann parent application, which interferences gave defendant access to the application of the patent in suit, as is shown by certified copies of the Patent Office records introduced in evidence by the defendant.

Defendant directs attention to a list of 18 patents, including the Kettering and Chryst and Osborne patents previously referred to, as illustrative of patents which have been or are in interference with the application of the patent in suit, or with divisions of that application, and in which the specific claims comprising the issues were made by Gubelmann more than two years after the issuance of the patent. Defendant, however, fails to point out any specific connection between these patents or interferences and the subject-matter in issue, other than what has already been referred to relative to the Kettering and Chryst and the Osborne interferences; so, therefore, these matters apparently have no direct bearing on the questions in issue.

Defendant also directs attention to a certain Carroll French patent, which issued in March, 1905, and a certain Carroll British patent, open for inspection in November, 1905, and the corresponding United States Carroll patent, No. 1,149,342, which was filed July 22, 1905, and issued August 10, 1915. It therefore appears that the United States patent was pending during and after the Spangler interference in 1905, at which time the Gubelmann application was directly before defendant, and that, if defendant had desired to contest these matters with Gubelmann in the Patent Office at that time, it might have done so. Instead, however, defendant proceeded to take out on August 10, 1915, its Carroll patent (filed July 22, 1905, 5½ years after Gubelmann filed), and claimed therein subject-matter to which Gubelmann had allowed claims, standing in his case in 1905 during the Spangler interference. I am unable to understand how defendant is in any position to urge equities in its favor in a situation, where it comes in late and makes claim to that which stood covered by allowed claims to another, who has preceded it 5½ years in the Patent Office. An example of a claim that stood allowed in the Gubelmann case, January 3, 1905, is as follows:

"42. The combination, with a recording mechanism, of a grand total registering mechanism, a subtotal registering mechanism, and a shifting mechanism, whereby said recording mechanism may be shifted relatively to either of said registering mechanisms, substantially as set forth."

This claim is clearly a claim to subject-matter of certain of the claims in issue — for example, claim 233, referred to above, which subject-matter defines an invention which is present in defendant's class 2000 machine.

Defendant introduced in evidence a machine that was identified by its witnesses, Lockyer and Lee, which machine was sent to France for exhibition in the year 1904. This machine is obviously too late to be prior art against Gubelmann, and in view of Gubelmann's claims, and the fact that the witness Lockyer testified that this machine was shown and described in the Carroll patent, No. 1,149,342, filed July 22, 1905, and issued August 10, 1915, this machine is not of any assistance to the defendant.

Finally, defendant selects the Piscicelli patent, No. 872,845, issued December 3, 1907, on an application filed December 11, 1903, and directs attention to claim 6 of this patent, which covers interspersed totalizers in combination with a kind of selecting mechanism specifically different from that disclosed in the Gubelmann application, and therefore it would not have been possible for Gubelmann to have copied this claim, even had he so desired. As we have seen, the application of this Piscicelli patent was filed long after Gubelmann filed, and was not acquired by the defendant until April, 1911, when defendant filed a reissue of the patent, after it had full knowledge of Gubelmann's inventions and claims through the Spangler interference. I am unable to see how this patent can be of any help to the defendant.

Various other patents were introduced in evidence by the defendant, on which it does not specifically rely, so I conclude that consideration or discussion of them is unnecessary.

Thus it appears that all of the patents and machines upon which defendant bases the defense of laches to defeat the full force and effect of the Gubelmann claims in issue, from covering inventions to be found in defendant's class 1700 and class 2000 machines, are of a date subsequent to the time when Gubelmann was asserting inventorship in the Patent Office in the application of the patent in suit to the subject-matter of the claims of this patent as issued.

In addition to the patents to Carroll, 1,118,103, issued November 24, 1914; Piscicelli, 872,845, issued December 3, 1907; and Carroll, 1,149,342, issued August 10, 1915 — the defendant took out three additional patents, to wit: Kettering, 1,109,763, issued September 8, 1914; Fassett, 1,132,711, issued March 23, 1915; Muzzy, 1,140,476, issued May 25, 1915, on which they rely as setting up some sort of an estoppel against Gubelmann. It appears, though, that all of these patents were applied for after the Spangler interference in 1904 and 1905, most of them having been applied for within a few years after that interference was declared, and after knowledge of the Gubelmann application came to defendant's attorneys. These applications, filed with this knowledge, issued with claims therein which counsel for defendant admit — even assert — are similar to those in issue in this suit, and read on the Gubelmann structure. How any equitable doctrine in favor of this defendant can be predicated on such a situation is beyond my understanding.

Here is an invention, the application for which was filed in the year 1900, thrown open in 1905 to the inspection of the defendant, whose attorneys then began to file applications in which they put claims to cover the earlier invention. It is indeed difficult to understand how the Patent Office permitted these patents to issue with such claims. However that may be, claims secured in such a manner cannot now be urged with good grace or conscience to defeat the prior rights of the applicant. Claims thus secured do not constitute intervening rights, but rather intervening wrongs. The whole situation presents a clear example of the abuses which are possible under the name of Patent Office procedure, and show how easily advantage may be taken of a meritorious, but impecunious, inventor, when opposed by learned counsel. To permit an estoppel to be raised in the name of equity for the benefit of the holders of patents so obtained would be a travesty on justice.

It seems clear that this is not a case where an inventor went into the Patent Office with a disclosure of two or more associated devices, and with claims directed to one group of mechanism, and emerged with a patent on claims to an entirely different mechanism than that originally claimed, and where the art in the meantime had grown up as to the later mechanism. The Gubelmann claims were always, from the beginning to the end, directed to features of his adding and recording machine, and defendant is chargeable with full knowledge of these inventions after the declaration of the Spangler interferences in 1905. Defendant took its chances, and proceeded to develop and market its class 1700 and class 2000 machines with knowledge of the Grubelmann application squarely before it. Defendant's conduct in attempting to secure claims to cover up the Gubelmann machine, and then to block the issue of the Gubelmann patent indicates, conclusively, I think, its appreciation of the effect of the Gubelmann patent when issued. The plaintiffs are well justified, in my opinion, in the assertion that the acts of the defendant in seeking to prevent the issuance of the patent in suit, are those of a conscious infringer.

The situation as to defendant's machines and defendant's patents, upon which it relies as an estoppel defense against Gubelmann, belong to the type which were referred to by Judge Rose in the case of Howe Machine Co. v. Coffield Motor Washer Co., 197 F. 541, at page 547, 117 C.C.A. 37, 43, in speaking for the Circuit Court of Appeals, Fourth Circuit, where he said: "But nothing which Howe or Ornold did in this matter gained for them anything which could without misuse of language be called rights. In the atmosphere of a court of equity it usually takes longer than 7 months and 16 days for intentional wrongs to ripen into intervening rights."

The facts in the case of Goodwin Film Camera Co. v. Eastman Kodak Co., 213 F. 231, on page 235, 129 C.C.A. 575, 579, apply to the case at bar, where Judge Coxe, speaking for the Circuit Court of Appeals, Second Circuit, said:

"The Eastman Company did not commence its experiments looking to the substitution of pyroxyline for paper as a film support until the latter part of 1888. These experiments were continued by Henry M. Reichenbach, the company's chemist, until February, 1889, and on April 9, 1889, he filed his application for a patent as assignor to the Eastman Dry Plate Film Company, and promptly received a patent dated December 10, 1889. During all this period, beginning with the experiments and ending with the patent, Goodwin's application was lying in the Patent Office. We are unable to see how the Reichenbach patent anticipates or limits the claims of the Goodwin patent. It may be that his process was an improvement on Goodwin, and that during the life of his patent he was entitled to the exclusive use of such improvement; but it can have no retroactive effect, it cannot destroy or limit an invention which was in esse before his invention was conceived.

"The defendant's brief, after alluding to the extensive manufacture of the Reichenbach film since 1889, continues as follows: `This is the industry which is enjoined by the decree of the court below, 7 years after the expiration of the Reichenbach patent 417,202, 24 years after the Eastman Company began the manufacture of nitrocellulose film, and 26 years after the filing of the application on which the Goodwin patent issued. And this industry is enjoined on a patent the application for which was uniformly rejected by the five different Examiners who successively had it in charge during its 11 years' pendency in the Patent Office.' Truly an extraordinary and deplorable condition of affairs! But who was to blame for it — Goodwin or the five Examiners who improperly deprived him of his rights during these 11 years? We are unable to see what he could have done to enforce his rights during this period, or how any blame can attach to him for his inaction."

So we may pertinently inquire here, who was to blame, the patentee, or the many Examiners who so persistently and for such a long time refused to allow the application with adequate claims, such claims as were ultimately allowed, or the patentee's opponents in the many interferences, or the Patent Office interference procedure, which was taken advantage of by counsel to delay opponent's application for some years and thus endeavor to wear him out, or shall we blame the defendant whose efforts succeeded in creating delays and the postponement of the issuance of the patent in suit?

It must be remembered, too, that the plaintiffs have property rights that are not to be lightly stricken down or impaired, merely because of events beyond the patentee's control. In Blandy v. Griffith, Fed. Cas. No. 1,529, 3 Fed. Cas. 675, Judge Swayne, in speaking of the property rights in a patent, said: "The rights secured by a patent for an invention or discovery are as much property as anything else, real or incorporeal. The titles by which they are held, like other titles, should not be overthrown upon doubts or objections capable of a reasonable and just solution in favor of their validity. This principle should be steadily borne in mind by those to whom is intrusted the administration of civil justice."

In Baltzley v. Spengler Loomis Mfg. Co., 262 F. 423, speaking for the Circuit Court of Appeals, Second Circuit, Judge Hough, at page 426, pointed out that "laches is not a mere matter of time, like limitation, but is a question of the inequity of enforcing the claim." In view of the facts at bar, the further statement of Judge Hough, at page 427, especially fits this case: "Under such circumstances, there is no inequity in regarding this infringement as persistence in wrongdoing, when the light was unusually strong."

Operativeness.

This brings us to a consideration of defendant's attack on the operativeness of the Gubelmann patent. This is an affirmative defense, and under it the defendant claims that the patent should be held invalid under the fundamental principle of patent law that an invention must be both new and useful, in order to be the subject of the grant of a valid patent.

The burden of proof with reference to this defense is heavily on the defendant, and as utility is one of the qualities necessary to patentability, the granting of the patent is prima facie evidence of it, and this is not negatived by the fact that the device is susceptible of improvement, or that like inventions are so far superior to it that they may entirely supersede the use of it. Under this defense the defendant must show, either that it is theoretically impossible for such a device to operate or demonstrate by clear proof that a person skilled in the particular art to which the invention relates has endeavored, in good faith, to make the patent work, and has been unable to do so, and it necessarily follows that such evidence is overthrown, if it is demonstrated by practical experiments made by credible persons that they have succeeded in producing the results claimed for the patent.

In support of this defense the defendant offered the testimony of Mr. Browne, its expert, and of Mr. Stoddard, an employee of its patent department. Their testimony was supported by models which were intended to illustrate the various defects claimed to exist in the mechanism disclosed in the patent in suit. The plaintiffs met this attack of alleged inoperativeness by calling their expert, Mr. Dyer, in rebuttal, who produced certain models of the mechanism of the patent in suit, as well as certain charts, for the purpose of demonstrating the fallacy of the defendant's contentions. He testified that the mechanism disclosed was operative. The defendant, on surrebuttal, did not call any of its witnesses to question the accuracy or correctness of plaintiffs' models, but contented itself with an attempt to demonstrate inoperativeness by a demonstration of its models as well as by arguments of counsel.

Defendant urges that there are faults in the disclosure of the mechanism of the Gubelmann patent which render the machine inoperative, and alleges that the most vital of these faults is to be found in the carrying mechanism for the subtotalizer. Defendant also urges defects in other parts of the Gubelmann disclosure, and it introduced in evidence four models, Defendant's Exhibits O, CCC, DDD, and NNN, to illustrate the features of its attack on the operativeness of the Gubelmann machine. It appears from the testimony that a great many other models have been built, but they were not produced. The models in evidence were built under the direction of defendant's witness Stoddard, who is employed by the defendant in its patent department, and has been engaged in writing specifications for patent applications. He did not work in the mechanical end of the business producing cash registers, in which branch of its business defendant employs several thousand hands. It does not appear from the testimony that the witness Stoddard ever produced any machine that was manufactured by the defendant or any one else, although it does appear that the defendant maintains development departments for this work.

Why should defendant give the job of producing the models above mentioned to an employee of its patent department, when it has regular departments for developing machines from inventors' disclosures? The correct answer, as it seems to me, is because these models were built under the direction of defendant's patent department, knowing that they were to be used as defenses against the Gubelmann patent, and were not constructed by defendant's production department, where there are employees who are especially skilled in producing operative machines, lest, using their skill, they might produce operative models representing the Gubelmann disclosure. This record abundantly shows that the patent department of the defendant company has, for years, been carrying on vigorous contests with Gubelmann in various interferences in the Patent Office, and on cross-examination the witness Stoddard admitted his familiarity, before he began to build these models, with certain of those interferences, and with the attack of alleged inoperativeness which was raised by Hopkins in the Hopkins v. Peters v. Gubelmann interference. Stoddard testified that he commenced in March, 1923, to build the principal model, Defendant's Exhibit NNN, which embodies all the features attacked as inoperative; that the first attempt at any operation of this model was in July or August, 1923; and that it was not completed until February, 1924. But the record shows that the defendant filed its answer in the present suit in May, 1923, and at that time it set up in its answer that the Gubelmann patent in suit "is for an inoperative, useless, and unpatentable structure."

Stoddard testified that during the construction of model NNN he conferred, at the factory of the defendant in Dayton, Ohio, with an attorney representing the Burroughs Adding Machine Company, and that, while model NNN was started before he saw the attorney, nevertheless changes in its construction were made subsequent to their conferences and exchange of views. The record also shows that Stoddard not only knew of the interferences between Gubelmann and the National Cash Register Company, but that he also knew about the interferences which involved the Burroughs Adding Machine Company. He also testified that the Burroughs Company had interferences with the Gubelmann applications, which are divisions of the patent in suit, and that these same questions of inoperativeness had been raised in those interferences. Therefore, if he had succeeded in building operative models, he would not only have defeated his employer's objects in this suit and in its interferences with Gubelmann, but he would also have defeated the contentions of defendant's licensee, the Burroughs Adding Machine Company, in their interferences.

It appears from Stoddard's testimony that the centers upon which defendant's models were laid out were not the same as the centers of the Gubelmann patent, and that he made, in the defendant's models, some 33 or 34 changes from the disclosure of the patent in suit. It also appears that, while tool makers and mechanics were used to build defendant's models, these men worked from drawings and layouts dimensioned and approved by Stoddard. I think the situation here fits what Mr. Justice Bradley said in Loom Co. v. Higgins, 105 U.S. 580, at page 586, 26 L. Ed. 1177: "When the question is whether a thing can be done or not, it is always easy to find persons ready to show how not to do it."

It is quite clear, from the testimony of the witness Stoddard, that he knew the points of inoperativeness which his models were supposed to illustrate when he started out to build the defendant's models. It is perfectly clear that defendant's models were built for the purpose of illustrating the alleged defects in the Gubelmann drawings, and not with any purpose or intention of helping or curing any deficiency that might exist in the mechanisms disclosed therein. We have here about the same situation as was present in Rynear Co. v. Evans (C.C.) 83 F. 696, and Judge Coxe, in characterizing it, said on page 697: "An experience of 14 years in patent litigation has convinced the court that, when an expert undertakes to prove that his adversary's process or machine is a failure, he always scores a success. It is much easier to make a machine that will not work than one that will."

Judge Hough, in A.B. Dick Co. v. Barnett (C.C.A.) 288 F. 799, quite aptly and succinctly expressed the thought which is here applicable when he said, on page 801: "We are of opinion that it is advisable to remember first how very easy it is not to do a thing, and how small is the step from not wanting to do a thing to declaring that it cannot be done."

It seems perfectly clear, to say the least, from all the surrounding facts and circumstances, that it was perfectly natural that Stoddard would have been desirous, as well as anxious, to demonstrate that the mechanism disclosed in the patent in suit would not operate. In other words, it was easier for him to show that it would not operate than that it would operate, and it is difficult to find a motive which would induce him to demonstrate that it would operate, for, no matter how honest he might have been as an individual, it would have been difficult for him, no matter how skilled, to design an operative mechanism, when we consider the knowledge he possessed of the constant attack made on the operativeness of the mechanism disclosed and the legal effect of a successful demonstration by models built under his supervision. It would be only natural that every suggestion favoring operation which might have come to his mind from his prior mechanical experience would necessarily have had to compete with his inclination to show inoperativeness.

It is a short step from not wanting to do a thing to not being able to do it. Things move along so rapidly nowadays that those who say it cannot be done are interrupted by some one who does it. The defendant company regularly employed some thousands of workmen, whose business it was and is to make operative cash registers for the market. Among these there are, no doubt, numerous engineers and others skilled in the business of making operative adding devices. If from this number a man of proper skill had been selected to make an operative model of the patent in suit, with the understanding that his employer's interests required the production of an operative model, or even with no knowledge of its intended use, and such a man had failed, proof of such facts would have been much more persuasive.

Stoddard was in court throughout the trial. He was examined at length. His answers to questions under cross-examination, as well as his demeanor on the witness stand, were not such as to inspire confidence in the credibility of his testimony. Some of his answers were evasive. His testimony, generally, created an impression that he knew exactly what the effect of a full reply would be on the issues of this feature of the case, and some of his answers were apparently carefully worded. For example, he professed to know very little about the interferences in the Patent Office between Gubelmann and the defendant, when under the facts disclosed by this record I conclude that he well knew all about those proceedings. His alleged reason for not knowing about them was predicated upon the alleged fact that he was too busy in writing specifications and doing special work to know much about the interferences. He testified that he did not know, even approximately, how many interferences were pending in any one of the six years he had been working in the patent department. But when confronted, on cross-examination, with an affidavit which he had made, and which referred to some 27 interferences, his attitude changed, and subsequently he admitted that he had copies of the affidavit with him on the stand.

After hearing such testimony on clear and definite questions of fact, it is difficult to give full credence to his testimony as to his mental attitude — as to whether or not he honestly endeavored to make operative models, which fact cannot be known or disproved, aside from Stoddard's testimony. A study of the models themselves shows that many things might readily have been done to favor operation, such as the artisan in this art would naturally do, which were left undone, all of which is convincing evidence as to the intelligence, skill, and willingness shown in their construction.

As it would have been difficult, even for a conscientious person, to have made an honest application of his ability and experience under the circumstances under which Stoddard worked, it is clear to me that the defendant's models cannot be considered as proof that the patent is inoperative, but that they are of importance here merely as illustrative models to show the points of inoperativeness which have been relied upon by the defendant, and for that purpose they are useful to demonstrate what the points are, but they certainly are not conclusive proof to show that the device of the patent in suit is inoperative. As the record shows, defendant's counsel recognized the importance of endeavoring to show Stoddard's "honesty of purpose" in the construction of the models.

As pointed out by Mr. Dyer, plaintiffs' expert, defendant's models are ingeniously constructed in features where no attack is made as to the Gubelmann disclosure, but, as to the parts attacked, where it comes to changing anything in aid of the specific construction disclosed by Gubelmann, no change is made.

It is the general rule that patent drawings are not required to be working drawings, but serve merely as an aid to convey the idea of the inventor, or as an aid to disclose the invention to those skilled in the art. Mr. Browne, for the defendant, testified on this point, as follows (Record, p. 694):

"X-Q. 212. The point I am getting at, Mr. Browne, is: Would you not testify from your experience, that Patent Office drawings are not working drawings, and are not expected to be? A. That is true.

"X-Q. 213. And would not you also testify that it is very doubtful, in the case of a patent on a complicated machine, that an exact reproduction would work, if the builder exactly followed the drawings of the patent? A. That would be unmistakably true.

"X-Q. 214. That is so? A. Of course, there are a lot of things that we do not expect draftsmen to get accurately.

"X-Q. 215. Is it not also a fact, and would not you also testify, that patents are directed to those skilled in the art, and in the reproducing of a patent structure they are expected to exercise their skill? A. Exactly.

"X-Q. 216. In your experience in the adding machine art, Mr. Browne, have you had occasion to consider a patent on an adding machine where, if the drawings were servilely followed, the machine would not operate? A. I think that would probably be true of every patent on an adding machine."

That Mr. Browne's experience is based upon settled rules of law is shown by a reference to the following cases: Crown Cork Seal Co. v. Aluminum Stopper Co., 108 F. 845 at page 849, 48 C.C.A. 72; Western Telephone Mfg. Co. v. American Electric Telephone Co., 131 F. 75 at page 77, 65 C.C.A. 313; Cutler-Hammer Mfg. Co. v. Union Electric Mfg. Co. (C.C.) 147 F. 266, at page 269; Manhattan Book Casing Mach. Co. v. E.C. Fuller Co., 204 F. 287, 122 C.C.A. 440; American Valve Meter Co. v. Fairbanks, Morse Co., 249 F. 234, at page 239, 161 C.C.A. 270.

In Hillard v. Fisher Book Typewriter Co. (C.C.) 151 F. 34, it was held that a patent may be perfectly good, even if no illustration accompanies the description contained in the specification. On page 42, Judge Ray said: "Whether or not the means for accomplishing a result covered by the patent in suit are illustrated is immaterial. The question is: Has the complainant in the patent in suit stated what his invention is, the result sought to be obtained, and has he described in the specifications the means for obtaining that result?"

It will be remembered that the Gubelmann inventions are of a broad nature in their general concept, and that, if Gubelmann had so desired, he might have drawn from the skill of this art, which, as we have seen, is complicated, for those mechanical structures in the art which had been developed prior to his inventions, without in any way destroying the patentable nature of his combinations here in suit. The general rule was stated by Mr. Justice Bradley in Loom Co. v. Higgins, supra, at page 591, where he said: "It may be laid down as a general rule, though perhaps not an invariable one, that if a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention."

However, Gubelmann chose to use mechanism of his own design, but in doing so he addressed his specification and drawings to those skilled in the art; that is, to those skilled in this highly technical calculating machine art, and not to the mere mechanical arts of the wagon maker or blacksmith.

The disclosure of an invention — that is, the specification of the patent taken with the drawings — must be sufficiently clear to enable a person skilled in the particular art to which the patent applies to make and use the invention. It is fundamental that the purpose of the disclosure in any patent is to enable the public to practice the invention — i.e., the idea of the invention — after the patent has expired. A member of the public, desiring to practice the invention after the expiration of the patent, approaches the problem of building a machine embodying the invention of the patent, not in a hostile attitude, but in a friendly manner. Under such circumstances we must assume that such person is trying to practice the invention for his own benefit, and so will give free rein to his mechanical abilities, and if he should find certain details shown in the drawings that will not work well, or not at all, if the drawings are followed slavishly, his desire to practice the invention will naturally impel him to muster all his mechanical experience, knowledge, and skill to correct the defects. It must be remembered that a mechanic skilled in the art under consideration must necessarily be a mechanic of a very high order. In fact, it is conceded by both sides that few, if any, arts are more complicated or highly technical in mechanics and mechanical operations than the adding machine and cash register arts. So that, when reference herein is made to a mechanic skilled in the art, we mean thereby a person who by education, development, training, and experience has become proficient in his understanding of the complicated mechanisms necessarily involved in the construction of calculating machines.

In the Loom Case, supra, Mr. Justice Bradley, discussing the subject-matter of the skill of the art, on page 585, said: "When an astronomer reports that a comet is to be seen with the telescope in the constellation of Auriga, in so many degrees of declination, and so many hours and minutes of right ascension, it is all Greek to the unskilled in science; but other astronomers will instantly direct their telescopes to the very point in the heavens where the stranger has made his entrance into our system. They understand the language of their brother scientist. If a mechanical engineer invents an improvement on any of the appendages of a steam engine, such as the valve gear, the condenser, the steam chest, the walking beam, the parallel motion, or what not, he is not obliged, in order to make himself understood, to describe the engine, nor the particular appendage to which the improvement refers, nor its mode of connection with the principal machine. These are already familiar to others skilled in that kind of machinery. He may begin at the point where his invention begins, and describe what he has made that is new, and what it replaces of the old. That which is common and well known is as if it were written out in the patent and delineated in the drawings."

Under this well-established rule of patent law, Gubelmann might have illustrated many of his constructions diagrammatically, and designated on these drawings totalizers, carrying mechanism, etc., and they would have been well understood by those skilled in this art.

In view of these fundamental observations, and in order to assure the inventor that his reward may be commensurate with the benefit his disclosure will give to the public, after it is free to practice his invention, the courts have always shown a liberal attitude in considering questions of adequacy of disclosure. The defendant's models were introduced in evidence during Mr. Browne's testimony, but it is important to observe, as the testimony shows, that Mr. Browne did not become active in the preparation of the case until March, 1924, and that defendant's large model NNN, covering all of the constructions which were attacked, was completed before that date, and Mr. Browne had no direction or supervision whatever with reference to that model. Defendant's complete theory on these points awaited Mr. Browne, when he began his study of the case.

During the testimony of Mr. Dyer, plaintiffs' expert, there were introduced in evidence five models, Plaintiffs' Exhibits 19, 20, 21, 22, and 24. These models clearly demonstrate the operativeness of the construction disclosed in the Gubelmann patent. All of the problems which defendant's models failed to perform were correctly demonstrated by plaintiffs' models. In fact, plaintiffs' large model, Exhibit 24, embodying all of the constructions in dispute, went far beyond any problems that were attempted on defendant's models, and correctly performed all of the problems submitted by defendant's counsel during the cross-examination of Mr. Dyer, who characterized those problems as "very cruel examples," which would "put any adding machine to the most severe test possible." Whether the plaintiffs' models fairly represent the invention of the patent in suit will now be considered.

It is quite obvious that the patentee had a clear appreciation of the problems which confronted him. When he decided to embody his idea of printing grand totals and subtotals from a plurality of totalizers in a machine having the two totalizers in interspersed relation, the wheels of one driving the other, he made it clear that, when the carry takes place in the grand totalizer, it must not, except in synchronous operations; take place in the subtotalizer until at a later time. He also made it clear that, when a grand totalizer wheel passes from 9 to 0 and a carry takes place in the next higher order grand wheel, a braking effect is to be impressed on the subtotal wheel of the order transferred to, in order to prevent that subtotalizer wheel participating in the carrying movement which takes place in the grand; i.e., the subtotal wheel is to be held back one step by a brake, until this step of motion is required at a later time in the operation. He not only explained fully the functions required, but he provided a form of mechanism to carry them out.

To illustrate the principle of the mechanism, the plaintiffs first produced a model (Exhibit 19), which was referred to as an electrical analogue, in order to show how the principle of the subtotal carrying mechanism could be carried out by using an electromagnet to operate the brake. True, this model was much more remote in construction than plaintiffs' other models, and so it is unnecessary to consider it further, except to say it was merely illustrative, and that counsel for defendant at oral argument conceded that this model would very likely be considered an equivalent of the Gubelmann mechanism. The patentee is not limited to the precise means, or mechanism, or form of device shown in his patent. He is certainly entitled, in view of the fact that he accomplished something entirely new in result, to a liberal range of mechanical equivalents. His invention was a very unusual one, and a very ingenious one, and it was indeed a difficult problem to solve, and he solved it.

Plaintiffs' model, Exhibit 20, illustrated the disclosure of the patent with the braking effect applied on a round part of the subtotal wheel. The application of a brake to a concentric part of any mechanism is the ordinary and well-known way of applying a brake. To apply a brake to a variable member is unusual, and I cannot escape the conclusion that any mechanic in this art or any other, if he found such construction troublesome, or that it caused difficulties in operation, would at least try the common and ordinary method of brake construction. The application of the brake on a concentric part of the subtotal unit is justified, and the defendant's criticism of plaintiffs' models, Exhibits 20 and 24, is not well taken.

Plaintiffs' model, Exhibit 20, is further criticized by the defendant, because the arm working on the units subtotal carrying cam is not identical with the arm working on the tens subtotal cam. While these arms look just alike in the patent drawings, it would seem to me that a mechanic would fashion these parts and others in such a way as to best adapt them for the work which it was intended by the patent they must perform. As pointed out in defendant's brief, the units' arm is merely a releasing arm. The tens' order in this model is a brake arm, although most of the braking is done by the brake pad, mounted on an arm pivoted to it. As their functions are not the same, it is only natural that they might be fashioned somewhat differently. Such changes flow naturally from the mechanic's desire to make the machine operate.

Defendant also objected to the fact that the arm which bears on the units' subtotal cam has its upper end bent forwardly, that it has an anti-friction roller where it bears on the cam, and that it carries a pin C, which seems to be merely a spacing pin. The defendant does not claim that these things have any particular function, and the plaintiffs' witness Dyer testified that he knew of none. They seem to be merely incidental details, that have nothing to do with whether or not the mechanism is inoperative in the manner claimed. Similarly I see no objection to attaching the springs lower down or higher up, or making one spring strong and another weak. A skilled mechanic will do things of that sort as a matter of course, in any faithful effort to make a device operate. Relying upon such trivial matters to prove the mechanism of the patent in suit inoperative, and what seems to me an unfaithful effort to eliminate these trivial objections and to produce operative models, shows a desperation to which the defendant has been driven upon the matter of its claim of alleged inoperativeness.

The patentee's contribution to the advancement of the art did not reside in such details, but in the broad combinations which he taught the art, and which have been put into extensive use by the National Cash Register Company and the Remington Cash Register Company. His contribution was not confined to such minor details as to whether or not all the arms 123 should be identical, and whether the braking function and the releasing functions should be combined into one arm.

Respecting plaintiffs' model, Exhibit 21, the defendant contends that the grand total and subtotal cams are of different shapes. This criticism is not justified, as Mr. Dyer pointed out that the tension arm 127 is caused to move a greater distance than the arm 123, due to the "kick" of the cam 57. A study of this model shows that the unbalanced effect referred to is due to the natural tendency of the tension arm 127 to move further by momentum than the arm which strikes it. In model 21 this is taken advantage of by constructing it so that the hook 133 will retain the tension arm 127 in the position to which it naturally moves by its own momentum when kicked forward by arm 129. Thus the tension arm is moved a greater distance than the brake arm, causing the desired unbalanced effect. It is not a fair characterization to call this an unbalancing of the cams by making them of different shape to such a degree as amounts to a disregard of the functions of the cams in total taking. In fact, the witness Dyer said that the effective functions of the cams are not changed. There is no substantial difference in shape between the cams, certainly no more so than in model, Exhibit 24, in which the plaintiffs admitted the cams were somewhat different in shape. Such difference in shape is immaterial, as the model operated correctly to position the type for total printing, both in the subtotalizer as well as in the grand totalizer. Therefore the construction of model 21 is not such as to ignore the total printing function of the cams.

Defendant objects to the fact that the cross head and releasing cam are not shaped exactly like those in the drawings. It is not expected that patent drawings be copied blindly. In any event the differences pointed out do not appear to have any bearing on the features of operation attacked. Plaintiffs' model, Exhibit 24, is a full-sized machine having two rows of amount keys. It has 3 grand totalizer wheels — units, tens, and hundreds. The brake mechanism for the hundreds subtotal wheel has not been applied in the model. Therefore the machine could add numbers totaling into the hundreds in both the subtotalizer and grand totalizer, except that the brake mechanism has not been applied to the subhundreds wheel. Therefore this wheel did not always show the correct total, and acted as the patentee said it would in the specification if no brake is provided.

However, this did not affect the accuracy of the machine within its capacity; i.e., the first two orders. In the problems used by the defendant to establish this defense, the error was shown to be in the tens order. The capacity of this model and of the others was sufficient to meet that attack. Plaintiffs' expert repeatedly said the model could have been built to carry further extensions, and the defendant did not deny this assertion. If plaintiffs had provided three orders in the model, then the defendant would probably have asked for four, and so on. Such objections cannot prevail. The model had sufficient capacity to show that the defects illustrated by the defendant's problems did not exist, and that the principle of the mechanism was not inoperative. They therefore served amply to meet the defendant's evidence on this point. While the defendant objects to various differences in details in this model, as in the others, they are not fundamental or substantial, and are clearly permissible. These, as well as those pointed out in connection with the other models, merely represent what any first-class mechanic in this art would do in a faithful effort to make the machine work. Thus the defendant objects to the use of a C-shaped spring on the end of the arm 54. This was provided, like a somewhat similar piece found in defendant's models DDD and NNN, to give greater flexibility to the end of the gooseneck arm. This construction is apparently not absolutely necessary, as it is not used in plaintiffs' model 22, which operates satisfactorily with a rigid gooseneck arm.

Plaintiffs' Exhibit 24 operated successfully for all the problems relied on by the defendant to establish inoperativeness. On cross-examination the defendant caused numerous problems to be worked on the machine, and correct results were recorded, except in one operation, when it subsequently appeared that a wrong key was depressed, as on repetition of the same problem the error was not repeated. But, even if this model, Plaintiffs' Exhibit 24, had failed to add correctly in this instance, that fact does not show inoperativeness, unless it appears that such misoperation is due to an error in principle. It is contended that an adding machine should operate correctly 100 per cent. of the time. To that proposition I subscribe, in the sense that its principle of operation must be such that, when properly carried out and in good working order, it must be capable of handling every possible combination of figures within its capacity. In McKenzie v. Cummings, 24 App. D.C. 137, which is cited and relied on by the defendant, it appears that the voting machine there under consideration erred as often as once in every 100 times. Judge Morris said on page 140:

"It is conceded by Cummings himself, in regard to these two machines, that they would not always operate, and that they were not entirely accurate. It was admitted that, as often as once in 100 times, the registering mechanism would fail to register as desired; and it is not clear from the testimony whether such inaccuracy was the result of some fundamental defect in the mechanism or merely of crude mechanical workmanship. If the defect was due merely to crude workmanship, it would seem that it would have been easy to show that fact. The sequel would appear to indicate that it was due rather to the absence of some necessary element; for accuracy was not attained until the construction of new and different mechanism in the successful machine of 1901."

It therefore appeared in that case that, when certain combination of conditions existed, the correct results could not be secured. Such results as the court said indicate the absence of some necessary element and a defect in principle — not merely a misoperation — which might have been due to the construction faults of the particular machine or model under consideration.

In Carlin v. Crumpton, 45 App. D.C. 166, also relied upon by the defendant, the plaintiff apparently admitted that the machine gave a wrong result when certain combinations of figures were introduced into the machine; but he contended that the application by the operator of certain rules, at irregular and uncertain intervals, would enable the operator to correct the error. It also appeared that the specification contained no suggestion that rules would have to be applied to reach the objective result. The Court of Appeals of the District of Columbia, speaking by Judge Van Orsdel, on pages 168 and 169, said:

"On final hearing, Crumpton again assailed the operativeness of Carlin's machine, whereupon Carlin set out certain rules the application of which by the operator would obviate the objection interposed and enable the desired result to be reached on his machine. * * * In other words, the machine of the issue is designed to make a correct calculation whether the result be a plus or minus quantity. It was intended to obviate the error of one unit in `passing through zero.' This cannot be accomplished by the Carlin machine directly `in an endless chain of carrying operations from first to last and from last to first,' as required by the counts. Before he can make the proper calculation for which the machine is designed, he must invoke his rules and make two other calculations, or, by observing closely the dials, add one additional unit under certain conditions.

"The failure of the Carlin machine to indicate correctly the true result in algebraic calculations is due fundamentally both to construction and theory of operation. The art to which this invention belongs is a highly technical one. The fact that the rules would only have to be applied at irregular intervals does not aid Carlin's contention. It only adds to the difficulty of successful operation. Without the intervention of the rules, the device is inoperative. With the rules its operation becomes so complex as to render it impractical when compared with the Crumpton machine."

That the difficulty was in the principle of the machine and not merely in construction faults is clear. The error above referred to, if any, may have been due to a poor adjustment of the parts of the model, but it was not due to an error in principle. As has already been noted, the mistake did not occur when the same problem was again introduced into the machine.

The Gubelmann machine is not predicated on a false "theory of operation," as was the case with the Carlin machine, nor is it inadequate to perform all its functions, due to the "absence of some necessary element," as in the case of the Cummings machine. Nor do I understand that there is any claim made that all machines operate correctly 100 per cent. of the time, as it is a matter of common knowledge that every machine will fail sometimes, and needs adjustment and repairs. Thousands of persons are daily engaged in keeping sewing machines, typewriters, telephones, electric appliances, automobiles, cash registers, adding machines, etc., in proper adjustment and running order, and it is necessary for large concerns to maintain "service" departments for the purpose of keeping all kinds of machines in proper repair, when they fail to perform their functions with certainty.

If we are to require adding machines to be so perfect that they will never fail to operate correctly, then the best commercial machines could not qualify as operative. Therefore, even if an error of operation did occur, this is not fatal, unless it appears that such error is due to a defect in the principle of operation adopted by the inventor. I conclude that there is no defect in the principle of the Gubelmann machine, and this record supports this conclusion.

In view of the fact that the mechanisms attacked as being inoperative are to be found in that part of the Gubelmann machine relating to the carrying mechanism, which is not an element in any claim in suit, no description was needed as to the questions heretofore considered. However, a brief description of the Gubelmann carrying mechanism and associated parts will now be necessary, in order to understand the features which are attacked as being inoperative.

When a carry takes place in the Gubelmann grand totalizer, as, for example, when the units order wheel has made one complete revolution, a member having a curved upper end, referred to as a gooseneck, drops from the high point to the low point of the unit grand totalizer snail cam, and the free end of this gooseneck strikes and swings a wing plate pivoted on the tens order gear segment. This swings out a flat carrying finger from above a control arm, and permits the tens order gear segment to descend an extra tooth, thereby turning the tens order adding wheel one step, thus performing the carry. Each gear segment of an order higher than units is provided with this plate and finger device. When the gear segments are restored to normal, the carrying fingers are automatically replaced, and the parts are again ready for the next normal adding operation.

When the grand totalizer and the subtotalizer both indicate the same amounts, carries are concurrent in both totalizers. But where the amounts in the totalizers are different (as where one indicates a grand total of a plurality of transactions and the other the subtotal of the last transaction), and a carry takes place in the grand totalizer, when no carry should occur in the subtotalizer, Gubelmann provides a braking mechanism, which holds back the carrying operation from erroneously turning the subtotalizer wheel until the time comes when a carry should take place in this wheel. In order to permit this operation, Gubelmann provides that the drive for each higher order subtotalizer wheel shall take place through a C-spring, so that, when such wheel is retarded by the brake, the turning energy is stored in expanding the C-spring. Now, when the next lower order subtotalizer wheel is turned one complete rotation, the brake is released and the delayed carry in the C-spring becomes effective to rotate the proper wheel one step, and the carry is now performed in the subtotalizer as it should be.

The mechanism for operating the brakes comprises latch arms, one of which is driven forward each time a gooseneck member drops from the high to the low part of a grand totalizer snail cam, when a carry takes place in the grand totalizer. This operation of the latch arm, which stretches a brake spring attached to a pivoted brake lever, and thus applies the brake, is brought about by a presser finger on the gooseneck member striking a cross-head on the swinging end of the latch arm, to drive the same forward to latched position. Now, when the next lower order subtotalizer wheel makes one complete rotation during subsequent adding operations, a lever carrying a releasing finger drops from the high to the low part of this lower order subtotalizer snail cam, and the releasing finger strikes the cross-head and releases the latch, so that the latch arm is free to relieve the tension on the brake spring, and thus the brake is released. As soon as this happens, the energy in the expanded C-spring, or, in other words, the energy of the delayed carry, is effective to turn the proper subtotalizer wheel one step, and the carry in the subtotalizer has been properly performed.

The defendant raised, by its pleadings, 11 separate grounds of inoperativeness set forth in defendant's notice to the plaintiffs, as shown by Plaintiffs' Exhibit 29. Five grounds of alleged inoperativeness were pointed out and discussed by Mr. Browne, and a sixth was raised by defendant's witness Stoddard, and defendant's brief adds an additional ground, and these will be discussed in their order.

The first ground is to the effect that premature braking will occur, and will cause the subtotalizer to register too little. This attack is based solely on the disadvantage of applying the brake on a variable brake drum, such as the cams 122. It is evident that the only function of the brake is to overcome the action of spring 99. All that the brake has to do is to put enough drag somewhere on the subtotalizer element, so as to collapse spring 99. The drag required to do this is constant, as pointed out by Mr. Dyer, and I agree with his explanation. Therefore the braking effect, which must be applied to the subtotal unit, should be constant. I think a mechanic would recognize that fact immediately, and that the application of the brake on the cam 122 would necessarily give a variable rather than a uniform effect. Therefore, if this variation, due to the different positions of the cam, should be found harmful or undesirable, a mechanic would call on his experience, and resort to the ordinary form of brake, and put it on some concentric part of the subtotal unit. To do this he might use another arm to carry the brake shoe, as plaintiffs have done in one of their models.

But defendant contends that the brake arm 123 has two functions in certain orders: (1) A braking function; and (2) a releasing function. If the arm which carries the brake should operate on a concentric surface, it would not have a falling movement which would permit it also to serve as a releasing device. It might, therefore, have been better to have used two arms — one to carry the brake, and the other to have the releasing function. This is the plan upon which certain of plaintiffs' models are built. Gubelmann provided one lever with two functions. Would it require invention to have these two functions performed by different levers? The patentee, to arrive at the means he illustrated in his patent, went further than to provide two levers. He proposed to save a part by combining two functions in one part. It does not seem to me that he should be penalized for going a step further than was necessary.

Under this alleged defect it was argued that the subtotal wheel would, under certain conditions, lose 1. Undoubtedly this was caused by the stops which had been placed in defendant's model, Exhibit CCC, and this conclusion is conceded to be correct by the witness Stoddard, as appears from his testimony, where the following questions were asked and answers given on pages 886 and 887 of the record:

"X-Q. 272. Can you answer this question yes or no: When you added 90 into the grand totalizer of this model CCC, and the subtotalizer stood at 80, was not the units lever 127, identified by that number in the Gubelmann patent, stopped by the stop for that lever that exists in the model? The witness: Yes; I would like to have it repeated. (X-Q. 272 repeated as recorded.) A. Yes.

"X-Q. 273. Now, that stop or that series of stops for the lever 127 are not in the Gubelmann patent, are they? A. No, sir."

The second ground of attack made by the defendant against the operativeness of the Gubelmann disclosure is based on the supposition that, under certain conditions, carries in the subtotalizer are not held delayed, and the subtotalizer will, at times, indicate too much. Defendant's proposition here is based on the supposition that, if there is a delayed carry in a subtotalizer member, and if additions in this member are sufficient to cause the brake arm to drop from the high to the low point on the subtotalizer snail cam, the friction of the brake arm against the snail cam, which friction during this movement is effective in an additive direction, acts as a drive, which, together with the delayed carry stored in the C-spring, furnishes sufficient energy to rotate this subtotalizer member one step. Defendant demonstrated this by its model, Exhibit CCC, in which leather friction material was added to the side of the brake arms to bear upon the snail cams. No such difficulty arose in connection with the operation of Plaintiffs' Exhibits 20, 21, and 24, which embodied mechanism including this feature of Gubelmann which is attacked. With certain of these models Mr. Dyer, plaintiffs' expert, correctly performed the identical problems which, when introduced into defendant's models, gave incorrect results.

The third ground deals with the grand total carrying mechanism. Defendant urges that under certain conditions the end of the gooseneck will not possess sufficient energy to swing the wing plate to withdraw the flat carrying finger from above the control arm, to permit the gear segment to descend to introduce a carry into a higher order grand totalizer wheel. Mr. Dyer pointed out that this is a trigger mechanism, which is nothing but a first-class lever, in which, the greater the leverage in favor of the power end of the lever, the less is the power required to operate it. And Mr. Browne, on cross-examination, admitted this to be true, as appears on page 715 of the record, where he testified as follows:

"X-Q. 366. Well, it is elemental mechanics, is it not, Mr. Browne, that, in the case of a first-class lever, the farther the work or weight from the fulcrum, the greater the power required? A. Yes."

In defendant's model DDD the wing plate, where the power is applied, is made quite narrow, while the flat finger over the control arm is made excessively long, and the parts are so spaced that the power applied works at a decided disadvantage. Furthermore, the end of the gooseneck is arranged to strike the wing plate relatively near the axis of its pivots, so that the application of the power is particularly unfavorable. Here, again, the fault lies, not with the Gubelmann disclosure, but with the construction of defendant's model. It was also observed that the wing plate in defendant's models was very hard to operate; in fact, it was impossible to move it, even when a considerable pressure of the finger was applied. This condition the plaintiffs claimed was due to the fact that the lug 50 was made much longer than necessary. On account of its great length, and its shape with respect to the surface 47, it tends to sweep rearwardly along the surface 47. This it cannot do, because the surface 47 rises. A mechanic working in this art would, I think, avoid that difficulty by making the lug 50 shorter and shape it so as to obviate this sweeping action. Nor does it require invention to overcome this trouble. Plaintiffs' models 22 and 24 show that the wing plate is easily tripped, and in the many operations of these two models, especially model 24, there was not a single instance when the gooseneck failed to trip the wing plate.

The fourth point of alleged inoperativeness raised by the defendant is in connection with the zero stops for the gear segments. Lugs are provided on the control arms for the gear segments, which set on shoulders on the ends of the differential stop arms, so that gear segments are held stationary, unless a key of that denominational order is operated. In defendant's models these zero stops cammed off and permitted the gear segments to improperly drop. It is perfectly obvious, if the faces of the lugs and the surfaces on the shoulders are inclined in such manner as to present camming surfaces when pressure is applied, as in the case in defendant's models, that the lugs will naturally cam the shoulders from beneath these lugs. It is equally apparent that, if these surfaces are not so inclined, pressure on the lugs will have no camming action whatever on the shoulders. This is so very obvious that an ordinary mechanic would be expected to make it right in the first instance; but, if the camming action should occur, the action itself immediately teaches the necessary change, namely, make the surfaces so they do not cam. Here, again, defendant's models depart from the teaching of the Gubelmann disclosure, and the fault lies, not in the Gubelmann patent, but with the defendant's models. No difficulty whatever was encountered on this point with plaintiffs' models, which do follow Gubelmann's exact disclosure as to these zero stops.

The fifth point, and the one raised by defendant's witness Stoddard, is that, during total taking operations and clearing the grand totalizer, when the tension on the springs which move the feeler fingers into engagement with the snail cams to control total printing is released, the return springs for the differential stop arms are of such strength as to disengage differential stops for the gear segments, so that the gear segments drop down to improper positions. Again the difficulty is with the improper construction of the differential stop faces in defendant's model, and the use of excessively strong springs for pulling back the differential stop arms. The fault is with the defendant's models, and not with the Gubelmann disclosure. Any mechanic skilled in this art, encountering this problem, would construct these stops properly, and would utilize springs of the proper strength to perform the required functions. This was done in plaintiffs' models, which demonstrate that these parts function correctly when properly constructed.

The sixth point of alleged inoperativeness, urged by defendant, is based on the complaint that the Gubelmann printing mechanism includes cross-rule type. Defendant claims that, since the cross-rule type are positioned by certain differential stops, the feeler fingers associated with these stops will contact with the spiral cams, and that the machine will print a total, instead of a cross-rule. This can hardly be sustained as a point of inoperativeness, even if this point be well taken, because all that is necessary to meet the defendant's objection is to clear the totalizer; then the cross-rule can be printed.

However, plaintiffs demonstrated by their models, Exhibits 22 and 24, that a cross-rule can be printed by the operation of the cross-rule key. Besides, the Gubelmann patent discloses a modified form of cross-ruling mechanism comprising a cross-rule bar, and defendant has not made any objection to the modified construction. Gubelmann's cross-rule mechanism, I therefore conclude, is clearly operative. In any event the matter is not in issue here, as none of the claims in suit includes the horizontal or cross-rule mechanism. This is merely an adjunct to the machine, and, even if it were inoperative, it would not interfere in any way with its adding functions, and, if desired, could be entirely omitted from the machine.

The additional point of attack on alleged inoperativeness, raised by defendant in its brief, has to do with an alleged loss in the subtotalizer, based on the supposition that, under certain conditions during the operation of the Gubelmann machine, the cross-head on the tension latch arms will get under the presser finger on the gooseneck, and after that time it never has a chance to again get in front of the end of the presser finger. Defendant uses this alleged difficulty as the excuse for introducing its adjustable stops to limit the rearward movement of the latch arms in defendant's models. This supposed difficulty is entirely overcome by the proper proportioning of the parts and the length of the brake springs. In plaintiffs' models, as previously noted, there are no stops for limiting the backward movement of the latch arms carrying the cross-heads, and in the operation of plaintiffs' models the cross-head properly comes in front of the presser finger at such time as it should, so that no difficulty whatever, on this point, will arise in a properly constructed device built in accordance with Gubelmann's disclosure.

Much has been said about the Hopkins v. Peters v. Gubelmann interference, and the whole record of that proceeding is in evidence as Defendant's Exhibit 46A. It is claimed that the Examiner of Interferences at one time held that the mechanism of the patent in suit was inoperative, and that this ruling has never been reversed on its merits. This decision was made by one H.E. Stauffer, who is now on the staff of the defendant as associate patent counsel, and he was present in court while this suit was tried. A reading of Mr. Stauffer's decision, which is found in the printed record of that interference from page 685 to page 716, shows at least one error in his reasoning, in that he assumed that there must be stops for the tension arms, which plaintiffs' models show are not necessary.

In any event, the evidence before Stauffer, as Examiner of Interferences, was not the same as that which was produced in court, and for that reason, if for no other, his conclusion is of no particular importance, and certainly of little weight here. Particularly is this true when the record shows that Stauffer merely ventured an opinion on a point which, as was found by the higher tribunals, including the Court of Appeals of the District of Columbia, was not properly before him. If any importance is to be attached to this, it is more than overcome by the fact that the Examiner must have been satisfied that the device was operative when he allowed the application.

The record further shows that Gubelmann appealed from that decision to the Board of Examiners-in-Chief, and based his appeal on the ground that the Examiner of Interferences should not have decided the matter urged by Hopkins, as in doing so he had violated his own rulings on the point and that of the Assistant Commissioner of Patents, and, further, because Gubelmann was not at that time prepared to argue the new features mentioned under features 5 and 6, and had not discussed them in his brief for final hearing, as a result of which "Gubelmann was taken utterly by surprise," as the interference record shows on page 742. This contention was evidently held to be well taken, as the Board reversed the decision of the Examiner of Interferences (page 786), on the ground that the matter was not properly before him, and this ruling was affirmed by the Commissioner of Patents (page 794), and by the Court of Appeals of the District of Columbia. Such a decision, made under such remarkable circumstances, by a minor official on a point not properly before him, and one on which he had no right to venture his opinion, and on evidence other than that before this court, is not such as to have any persuasive effect in this suit, to say the least.

In the case at bar we have had the benefit of the testimony of able expert witnesses, and of facts brought out on cross-examination, as well as the further advantage of observing demonstrations on models produced by defendant and models produced by plaintiffs. This testimony and these demonstrations leave in my mind no doubt as to the operativeness of the disclosure of the Gubelmann patent, and show how easily a purely theoretical argument may lead to erroneous conclusions. In Packard v. Lacing-Stud Co., 70 F. 66, at page 67, 16 C.C.A. 639, 640, Judge Putnam, speaking for the Circuit Court of Appeals, First Circuit, said: "Persons possessed of the most brilliant conceptions are sometimes the poorest mechanics."

The Bell Telephone patent ( 126 U.S. 1, 8 S. Ct. 778, 31 L. Ed. 863) was attacked because it was alleged to be inoperative, and in discussing that attack at length Chief Justice Waite said, at page 535 ( 8 S. Ct. 782):

"It is quite true that, when Bell applied for his patent, he had never actually transmitted telegraphically spoken words, so that they could be distinctly heard and understood at the receiving end of his line; but in his specification he did describe accurately and with admirable clearness his process, that is to say, the exact electrical condition that must be created to accomplish his purpose, and he also described, with sufficient precision to enable one of ordinary skill in such matters to make it, a form of apparatus which, if used in the way pointed out, would produce the required effect, receive the words, and carry them to and deliver them at the appointed place. The particular instrument which he had and which he used in his experiments did not, under the circumstances in which it was tried, reproduce the words spoken, so that they could be clearly understood; but the proof is abundant, and of the most convincing character, that other instruments, carefully constructed and made exactly in accordance with the specification, without any additions whatever, have operated and will operate successfully. A good mechanic, of proper skill in matters of the kind, can take the patent, and, by following the specification strictly, can, without more, construct an apparatus which, when used in the way pointed out, will do all that it is claimed the method or process will do. Some witnesses have testified that they were unable to do it. This shows that they, with the particular apparatus they had and the skill they employed in its use, were not successful; not that others, with another apparatus, perhaps more carefully constructed or more skillfully applied, would necessarily fail. * * * If one succeeds, that is enough, no matter how many others fail. The opposite results will show that in the one case the apparatus used was properly made, carefully adjusted, with a knowledge of what was required, and skillfully used, and that in the others it was not."

Imperfections in a machine, not affecting the substance of the invention, and which are curable by mechanical skill, do not render the patent inoperative. Dalton Adding Mach. Co. v. Rockford Milling Mach. Co. (D.C.) 253 F. 187; Crown Cork Seal Co. v. Aluminum Stopper Co., 108 F. 845, 48 C.C.A. 72; Western Telephone Mfg. Co. v. American Electric Telephone Co., 131 F. 75, 65 C.C.A. 313; Engineer Co. v. Hotel Astor (D.C.) 226 F. 779; Electric Smelting Aluminum Co. v. Pittsburg Reduction Co., 125 F. 926, 60 C.C.A. 636; Scott v. Fisher Knitting Mach. Co., 145 F. 915, 76 C.C.A. 447; Carnegie Steel Co. v. Cambria Iron Co., 185 U.S. 403, 22 S. Ct. 698, 46 L. Ed. 968; Dashiell v. Grosvenor, 162 U.S. 425, 16 S. Ct. 805, 40 L. Ed. 1025.

As every patent has attached to it the presumption of validity, this presumption carries with it the presumption of operativeness, and in support of this rule it is sufficient to quote what Mr. Justice Brown said in Dashiell v. Grosvenor, supra. On page 432 ( 16 S. Ct. 807) he said: "The very fact that a machine is patented is some evidence of its operativeness, as well as of its utility, and where a model is constructed after the design shown in a patent which is not perfectly operative, but can be made so by a slight alteration, the inference is that there was an error in working out the drawings, and not that the patentee deliberately took out a patent for an inoperative device."

The presumption of operativeness is much stronger in the case of a broad, or generic, or primary invention, such as the patent now under consideration. Such an invention should not be stricken down and declared invalid, because inoperative, without ample, satisfactory, and convincing proof. The necessity for the correction of minor structural defects in a machine does not necessarily establish a false inventive concept, and the fact that certain changes, within the realm of the skill of a mechanic experienced in this art, are necessary to successful operation, will not support a claim that the patent is void for inoperativeness.

In Engineer Co. v. Hotel Astor (D.C.) 226 F. 779, Judge Hough, on page 783, said: "A device need not be perfect in order to escape the charge of inoperativeness. * * * The test of operativeness is to ascertain whether the patented device does (even lamely and imperfectly) perform the acts claimed for it in the method described and (perhaps) for the reasons given." Affirmed 226 F. 949, 141 C.C.A. 553.

Judge Townsend, speaking for the Circuit Court of Appeals of the Second Circuit, in Scott et al. v. Fisher Knitting Machine Co. et al., 145 F. 915, on page 922, 76 C.C.A. 447, 454, said: "It is frequently a characteristic of generic inventions that their first embodiments work imperfectly, and where the imperfections may be remedied, as in this case, by what amounts to a mere readjustment of relative sizes, such change does not affect the character of the underlying creative conception."

We must not lose sight of the fact that the invention disclosed by the patent in suit was presented in 1900, 25 years ago, when it may fairly be said that the art was then in its infancy; for this was the first time a mechanical accounting machine was invented that would record the items and print subtotals of a series of bills, and at the same time preserve the grand total. The record shows that the art, prior to the application of the patent in suit, had not developed much beyond a machine which could print items and then a total of those items. Up to 1900 it was not known how to print items and a subtotal for one person's purchase, then clear the subtotalizers, so that it was ready to print the items and a subtotal for the second person's purchase, while at the same time preserving a grand total of both the first and second person's purchases, and then print the grand total. "The underlying creative conception" of the patent in suit lies in the invention of the grand and subtotalizers with printing mechanism to print items and subtotals and automatically clear the subtotalizer after the subtotal had been printed, and of having a common printing mechanism that printed items under key control and printed totals under control of either the grand totalizer or subtotalizer, and of mounting these totalizers in the most efficient way with the plurality of totalizers interspersed and co-operating with selective mechanism capable of printing from any one of the interspersed totalizers. If all these operations could be accomplished successfully, it seems to me that it is very reasonable to hold that the patent in suit discloses a generic invention, especially in view of the state of the prior art, as shown by this record, and as conceded by Mr. Browne, to which reference has been made in the early part of this opinion.

From all of the above, as well as from the disclosure in the patent in suit and the specification, it seems to me that this record very amply justifies the conclusions that we are here dealing with, and considering a generic invention, and that all discussions concerning it must be based upon that dominant feature, and all rules to be applied must be the rules which govern when considering such a patent, and not the rules applicable to a simple or improvement patent.

Nor is it necessary, in such an instance, to show that the device is a commercial success, for the Supreme Court so held in Hildreth v. Mastoras, 257 U.S. 27, 42 S. Ct. 20, 66 L. Ed. 112. Mr. Chief Justice Taft, speaking for that court, on page 34 ( 42 S. Ct. 23), said: "It is not necessary, in order to sustain a generic patent, to show that the device is a commercial success. The machine patented may be imperfect in its operation; but if it embodies the generic principle, and works, that is, if it actually and mechanically performs, though only in a crude way, the important function by which it makes the substantial change claimed for it in the art, it is enough. Telephone Cases, 126 U.S. 1, 535 [ 8 S. Ct. 778, 31 L. Ed. 863]; Mergenthaler Linotype Co. v. Press Publishing Co. [C.C.] 57 F. 502, 505."

In the Mergenthaler Case, just cited by the Chief Justice, Judge Coxe, on page 505, said: "The proposition upon which they appear to lay the greatest stress is that neither patent describes or claims an operative machine because neither is capable of `perfect justification,' viz.: Making lines of exactly the same length. Their contention proceeds upon the untenable proposition that the machine which produced the `linotype' was valueless because it did not produce an absolutely perfect `linotype.' Such a proposition, if sustained, would condemn to obscurity some of the greatest works of human genius. A great poem may be marred because the meter halts at times, but it is a great poem still. Even the masterpiece of Rubens was improved by the touch of his pupil, Van Dyck. * * * Concede that the machine when first produced was not perfect and that to Schuckers belongs the credit of producing the spacers which made it perfect. Cui bono? It would certainly be a novel doctrine to deny to an inventor the fruits of a broad invention because the machine which first embodied it was rudimentary in character and failed to do as good work as improved machines made subsequently. None of the great inventions could survive such a test. Ten years after the invention of Howe, the machine first made by him would hardly have satisfied the least exacting sewing woman. The Dodds and Stephenson locomotive would, only a short time after its construction, have been discarded as behind the age even by the savages of Tasmania. The telephone of Bell is not the perfected telephone of commerce, the Morse telegraph is looked upon to-day as an interesting antique. And yet, it would be an unheard of proposition to withhold from these illustrious men the credit they deserve because their machines were crude at first and were improved afterwards."

In A.B. Dick Co. v. Barnett, 288 F. 799, Judge Hough, speaking for the Circuit Court of Appeals, Second Circuit, on page 801, said: "We further think that, in judging the sufficiency of a patent disclosure, it is necessary to consider, not only the art to which the invention may be assigned, but its relation to other and especially older arts, and the nature of the alleged new thought. It is upon the difficulty, obscurity, or even novelty of the art concerned that depends the kind and degree of skill or knowledge which must be possessed by those who assume to apply it. * * * It is not possible to formulate a rule in this matter, though it may be safely said that it is by no means enough to condemn a patent disclosure that even a skilled man might find it necessary to make several tests or trials of before arriving at success. Malignani v. Jasper, etc., Co. (C.C.) 180 F. 442. The nearest approach to a rule is the summary declaration so characteristic of the rude sense of Jessel, M.R., that `the specification of a patent is not addressed to people who are ignorant about the subject-matter. It is addressed to people who know something about it.' Plimpton v. Malcolm, 3 Ch. Div. 531."

Nor is there any conflict in the law propounded in the cases just cited with the claims of law made by the defendant upon this feature of the case, and as set forth in cases cited by the defendant. For instance, reliance is placed upon Judge Hand's decision in Manhattan Book Casing Mach. Co. v. E.C. Fuller Co. (C.C.) 274 F. 964. As pointed out by counsel for plaintiff at final argument, the facts in that case were very different from the facts in this record. There the defendant presented proof that the machine of the patent was inoperative, and the plaintiff did not attempt in rebuttal to meet the proof thus adduced. This appears quite clearly when the entire opinion is read. For instance, Judge Hand there said, at pages 966 and 967:

"Now, it is here in evidence with no contradiction at all, as I shall show later, that the machine actually disclosed by the patent was inoperable. If the faults were merely in the diagrammatic representations, that would not be a defense, for diagrams or descriptions are not meant for working drawings. Crown Cork Seal Co. v. Aluminum Stopper Co., 108 F. 845, 48 C.C.A. 72. But such drawings must at least show enough for the ordinary skilled mechanic to build the machine, correcting the discrepancies and the difficulties appearing in the drawings, by means of that store of expert information which such men will bring to the task in a bona fide effort to make the machine work. * * * It must therefore be conceded, and indeed it is not denied, except as I shall show later, that the machine as disclosed was an inoperable mechanism. However, the defendant must go further, and show that it could not be made operable by a skilled mechanic. This it does by the testimony of its two experts, who say that they do not see how it could be made to operate."

The court then goes on to say (page 968) that it is not enough for plaintiff's witness (defendant's experts) "to say merely that he disagreed with them in their subsequent conclusion that it could not be made workable. * * * I must conclude that he would not venture to show in detail how the difficulties could be remedied." That that case was decided on the weight of the evidence adduced, or rather, on the lack of evidence offered by the plaintiff is also clear from the decision of the Circuit Court of Appeals, affirming Judge Hand, in 204 F. 286, 122 C.C.A. 440.

Nor is there anything inconsistent in Judge Denison's opinion in Computing Scale Co. v. Standard Computing Scale Co., 195 F. 508, 115 C.C.A. 418, cited and relied upon by defendant, with the general rules herein stated, and which are applicable to the questions here under consideration. In fact, the rules therein given are concisely stated, and are the very ones here applied to the question of alleged inoperativeness.

And finally, as it seems to me, it does not lie in the mouth of an infringer to deny the utility of a patent. The presumptions are, at least, against him. Moreover, if an infringer has seen fit to depart from the many articles and methods open to his use, and to adopt that of the patentee, there is a strong indication that the patent marks a distinct and useful advance in the progress of the art. Salt's Textile Mfg. Co. v. Tingue Mfg. Co. (D.C.) 227 F. 115; Electric Smelting Aluminum Co. v. Pittsburg Reduction Co., 125 F. 926, 60 C.C.A. 636; Lehnbeuter et al. v. Holthaus, 105 U.S. 94, 26 L. Ed. 939; Du Bois v. Kirk, 158 U.S. 58, 15 S. Ct. 729, 39 L. Ed. 895. And so, after a very careful and painstaking study of the great mass of documentary evidence introduced, and a very careful scrutiny of the models and machines in evidence in connection with the defendant's attack on the operativeness of the Gubelmann disclosure, I am convinced that the Gubelmann patent discloses an operative machine and that the defense of alleged inoperativeness must fail.

Failure to Call Gubelmann as a Witness.

The defendant complains that Gubelmann was not called to the witness stand to testify concerning matters which it alleges were of importance, and which only Gubelmann could explain, and asks the court to draw unfavorable inferences against such failure, under the well-known doctrine that, where a party omits to produce that evidence in elucidation of the subject-matter which is within his power, and which rests within his knowledge, this affords presumptions against him and raises strong suspicion that such evidence, if adduced, would operate to his prejudice. But the defendant overlooks the other well-settled rule that, where an adversary, in the proof adduced in support of his contentions, offers evidence which counsel on the other side are satisfied proves matters for which he is contending, it is an idle ceremony to call a witness to prove the very matters which have been introduced by his adversary, and which substantiate his claims.

After the defendant rested, counsel for plaintiffs very properly, as I think, did not call Gubelmann, because the features of their case, which Gubelmann could prove, had already been proven by defendant's evidence. By this procedure counsel for plaintiffs avoided an incumbrance of a record already sufficiently heavy, in which the evidence was present upon which to base their argument. So far as proof respecting the models was concerned, the plaintiffs had complied with the rules, and to have called Gubelmann would have resulted in producing merely cumulative evidence, which was not required, as in a certain sense the models spoke for themselves, and required no further demonstration than that which had been given by plaintiffs' other witnesses. The evidence was in on both sides, from which counsel could argue, and from which the court would have to decide.

The last five pages of the defendant's brief are devoted to a discussion as to the admissibility of certain file wrappers and other papers therein enumerated. At trial they were all marked for identification, subject to final ruling in this opinion. It is unnecessary to discuss the questions of law which defendant claims govern their admissibility, as I deem it advisable, under all the circumstances of this unusual case, that they be admitted.

The patents to Carney, No. 699,100, issued April 29, 1902; Cleal, No. 718,565, issued January 13, 1903; Carney, No. 748,259, issued December 29, 1903; Osborne, No. 1,105,456, issued January 23, 1912; and McCormick, No. 1,103,068, issued July 14, 1914 — have all been carefully considered, and the claims in suit are nevertheless found to be valid over those patents. Defendant directs attention to the Gubelmann divisional applications and other papers, showing the meaning of the word "indicators." At trial these were offered as Defendant's Exhibits 25A, 27A, 55A, 56A, 57A, 58A, 59A, and 60A for identification. While the plaintiffs claim that these documents are not admissible, I have concluded that it is advisable to admit them, so that all the facts may fully appear in the record. The corresponding papers, offered by the plaintiffs as Exhibits 26 and 28, for identification, are also admitted. For the same reason the documents marked Exhibits 26A, 52A, and 61A for identification, together with Plaintiffs' Exhibit 27 for identification, are also admitted.

The matters to be here determined and the questions to be answered are three: (1) Is the Gubelmann patent valid? (2) Are the defendant's machines 1700 and 2000 infringements thereof? And (3) are there any bars which limit the full force and effect of the Gubelmann patent? Upon the authorities cited, and for the reasons given, the first two questions are answered in the affirmative, and the third in the negative.

Such being my conclusions, it follows that the 10 claims of the Gubelmann patent relied upon as covering the defendant's class 1700 machine, to wit, claims 58, 62, 213, 214, 289, 290, 292, 294, 296, and 298, are held to be valid and infringed, and the 38 claims relied upon as covering the defendant's class 2000 machine, to wit, claims 10, 25, 28, 50, 52, 54, 55, 58, 60, 61, 62, 81, 92, 123, 127, 146, 147, 160, 200, 213, 214, 223, 224, 227, 233, 245, 247, 249, 250, 251, 252, 253, 254, 257, 282, 283, 285, and 287, are valid and are infringed thereby. It follows, therefore, that there may be a decree for plaintiffs, with costs to abide the event; and it is so ordered.