SANBORN, District Judge.
Suit for infringement of the De Vil-biss patent No. 12,137, for a computing scale, called the “Toledo scale,” reissued July 28, 1903, application filed November 16, 1901. The original is No. 649,915, dated May 22, 1900. The defenses are invalidity of reissue, want of novelty, and no infringement. The reissue is claimed invalid because alleged to be for a different subject-matter than the original, no inadvertence or mistake other than a possible error of judgment in accepting the original claims, and that the original patent record expressly concedes the invalidity of the subject-matter. More in detail, it is urged in defense that the only mistake was an error of judgment in acquiescing in the various rulings of the ‘Patent Office as to divisions and questions of aggregation, and in purposely conceding all claims relating to computing scales in order to obtain a patent on a form of disk bearing. Defendant’s machine is referred to here as the “Moneyweight” scale, and plaintiff’s as the “Toledo.”
Can the reissue be sustained with the rule of liberal construction applied, consistently with the decisions of the federal courts? Topliff v. Topliff, 145 U. S. 156, 12 Sup. Ct. 825, 36 L. Ed. 658; Parker & Whipple Co. v. Yale Clock Co., 123 U. S. 87, 8 Sup. Ct. 38, 31 L. Ed. 100; McCormick v. Aultman, 169 U. S. 606, 18 Sup. Ct. 443, 42 L. Ed. 875; Mahn v. Harwood, 112 U. S. 354, 5 Sup. Ct. 174, 6 Sup. Ct. 451, 28 L. Ed. 665; Corbin, etc., Co. v. Eagle Lock Co., 150 U. S. 38, 14 Sup. Ct. 28, 37 L. Ed. 989; Dobson v. Lees, 137 U. S. 258, 11 Sup. Ct. 71, 34 L. Ed. 652; Miller v. Brass Co., 104 U. S. 350, 26 L. Ed. 783; Franklin Co. v. Illinois, 138 Fed. 58, 70 C. C. A. 484 (in this circuit). While the statute does not limit the time within which reissue must be applied for, yet it is settled that the patentee must act promptly upon discovery of the error (Ives v. Sargent, 119 U. S. 652, 7 Sup. Ct. 436, 30 L. Ed. 544), especially after public use of the device covered by the reissue claims, or after such an act as would have been infringement if the reissue claims had been in the original grant.
In this case there was no delay after the patentee discovered the alleged mistakes, nor any public use or infringement prior to reissue. The question is thus narrowed to an inquiry whether there was such inadvertence or mistake as to authorize a reissue. Certain mistakes, even though present, are not enough. Examples of these are mis[559]*559takes creating public rights acted on, or arousing an estoppel in favor of another patentee, as in Dobson v. Lees, and mistakes attended with negligence or laches. The first two examples are not found in this record, because there was no public use, nor any estoppel, nor delay; but it is insisted that the patentee abandoned his invention by canceling all his original claims relating to computing scales.
As to the finality of the decision of the Commissioner finding inadvertence or mistake, the earlier rule was that the granting of a reissue closed all inquiry into the existence of inadvertence, accident, or mistake, unless it appeared that the Commissioner had no jurisdiction, having exceeded his authority. Seymour v. Osborne, 11 Wall. 516, 20 L. Ed. 33. • But this has been modified, in harmony with that applying to other executive action. The rule now is that the question whether the whole record shows inadvertence or mistake is one of law, which will be reviewed by the courts. While inadvertence and diligence are usually mixed questions of fact and law, yet as all the facts are generally of record, without dispute or conflict, the proper deduction is a conclusion of law. This agrees with the. almost universal rule governing executive decisions, which are, except in cases of deportation of aliens, reviewable for mistake of law, but not of fact, or mixed law and fact. United States v. California, etc., Co., 148 U. S. 31, 43, 13 Sup. Ct. 458, 37 L. Ed. 360. In Huber v. N. O. Nelson Mfg. Co., 148 U. S. 270, 33 Sup. Ct. 603, 37 L. Ed. 447, it was held, approving the rule adopted by the Circuit Court, that, where all the evidence presented to the Patent Office by the patentee applying for reissue is offered in an infringement suit, the court could review the finding at least to the extent of determining whether, as a matter of law, what was alleged to be a mistake was such as warranted a reissue. Franklin v. Illinois, supra ; Featherstone v. Bidwell, 57 Fed. 631, 6 C. C. A. 487. The matter was thus “manifest from the record,” according to the rule laid down in Topliff v. Topliff.
Mistake or inadvertence making a patent inoperative or invalid may be that of the patentee or his solicitor, either in preparing the specification or claims. If the solicitor fails to understand and properly describe or claim the real invention, by making claims so broad as to be anticipated, or so narrow- as to be inoperative, this is such mistake as to authorize a reissue, if authorized on other grounds. Topliff v. Topliff. This kind of mistake is the one particularly claimed by complainant, although the specification is also alleged to be imperfect.
The. computing scale in question is clearly shown in the drawings, which appear later, and are the same in the original and reissue applications, except as to certain extra features abandoned. It is not complicated, though multifarious, including some 46 elements, all found, in some form, in the prior art. It is used chiefly in grocery stores, and has had a very large sale. Some 3,000 were marketed even before the reissue was applied for. When it is desired simply to ascertain the cost of an article sold at a certain price by the pound, it is placed on the scale, and the index hand swings to the right and stops at a place on the price computing table at the top of the machine indicating the pounds and fraction of a pound which the article weighs. [560]*560The clerk then looks down the right-hand side of the index hand until his eye rests upon the price per pound, when he sees the total price in figures immediately adjoining and slightly below the point where the pound price appears. If a dollar’s worth of sugar at six cents a pound is desired, he puts enough sugar on the scale to make the index swing to the dollar mark indicated opposite the price per pound. And if a certain weight or valued quantity is wanted in a jar or basket, he places the receptacle on the scale and moves the tare weight to the right until the scale balances, then going through precisely the same operation as before. The operations are rapid and accurate. What the adding machine does in a bank the computing scale does in a grocery.
The Toledo reissue does not fall within the express terms of the statute relating to reissues, being section 4916, Rev. St. (U. S. Comp. St. 1901, p. 3393), because the claims were broadened. It is there provided that whenever a patent is inoperative or invalid by reason of a defective or insufficient specification, or by reason of the patentee claiming as his own invention or discovery more than he had the right to claim as new, if the error has arisen by inadvertence, accident, or mistake, and without any fraudulent or deceptive intention, a new patent may be issued for the balance of the term, on surrender of the original. Now the main difficulty with the original patent was in taking claims much too narrow; not in claiming more than the patentee was entitled to. So that the statute does not expressly cover the case. The general power of reissue, however, does not depend on statute, but on the general spirit and purpose of the Constitution, and the laws relating to patents. This was held by the Supreme Court in 1831, in an opinion by Chief Justice Marshall, in the case of Grant v. Raymond, 6 Pet. 218, 8 L. Ed. 376. And it is thoroughly established, since the passage of section 4916 in 1836, that a reissue may broaden a claim in a proper case, as well as narrow it. Topliff v. Topliff. For a time patents reissued with broadened claims were quite obnoxious to the courts, on accoitnt of the fraud and injustice connected with the reissue proceedings. The later tendency, however, is much more favorable. The law encourages honest reissues, condemning only those which are fraudulent, unjust, or negligent. Topliff v. Topliff; Eames v. Andrews, 122 U. S. 40, 7 Sup. Ct. 1073, 30 L. Ed. 1046; Peoria v. Cleveland, 58 Fed. 227, 7 C. C. A. 197; Houghton v. Whitin, 153 Fed. 740, 83 C. C. A. 84; Mathews v. Flower (C. C.) 25 Fed. 830, Brown, Circuit Judge. The question, therefore, is whether the Toledo reissue was fair and just, giving to the patentee the invention, and only the invention, which he actually made, but failed through mistake to obtain.
Mistake in the specification is claimed; but while it is crude and inartistic, lacking clearness and coherence, yet it is possible, by aid of the drawings, to understand how the machine works. Whatever were its defects and errors, they were not, as I think, the proximate cause of the inoperation and practical uselessness of the original patent. The claims, on the other-hands are of a character far different.
Twenty-nine claims, in all, were drawn by- the solicitors for the pat-[561]*561entee in tlie original proceedings, 11 of them in the application. Inasmuch as the prior art covered substantially all the elements of the combination which constituted the patentee's sole claim to invention, it was the business of the solicitors to claim the combination as a unitary structure, and not the specific things going to make it up, as well as to avoid broadening the individual elements by claiming means for the operation of the parts, instead of the particular devices shown in the drawings; taking care, however, not to narrow them further than necessary to escape the references. These things were accomplished fully in the reissue claims. But the solicitors, for some reason, in the original application inadvertently failed to properly claim the real invention. In the whole 29 claims formulated there is not one properly drawn. Perhaps the'nearest approach to a good claim was the twenty-fifth, being the third claim under date of February 2, 1900, which follows:
(25) “8. In a computing scale, the combination with the beam mounted on a fulcrum, the pan or platform, and a link; of the housing having a table provided with a single row of weight units and several rows of price totals, a normally vertical pendulum and normally oblique index hung on a common central pivot, the index having several price per pound figures, coacting with the price-total rows on the table, a link-pivot rigid with and normally above said central pivot and with which said link is connected means for counterbalancing the link-pivot and link, and a screw and nut opposite to and adapted to counterbalance the index, as and for the purpose set forth.”
Not only are the tare device and dash-pot left out of this claim, but the position of the link on the beam is not indicated. Where is the fulcrum of the beam, and how supported ? These defects are not supplied by any other claim then on foot. Very many of the 29 claims do not cover the computing feature, and most of the others are so broad in their details, and cover so few of the elements, that they were clearly objectionable. I think it appears with the greatest clearness that the proximate cause of the final omission of all claims to a computing scale, and the acceptance of two narrow claims to certain specific features of a noncomputing scale, was the inability of the solicitors to put the claims in proper form. Further, I think it is equally clear that, inasmuch as the solicitors never claimed the real invention, the abandonment of what they did claim was not a giving up of the invention ; especially as neither the public nor any alleged infringer used the invention prior to the reissue. If I am right in these inferences, then the inadvertence inducing the inoperativeness and invalidity of the first patent was such inability. There was no mistake of fact, hut the situation is accurately described hv the broad term, inadvertence, oversight, inattention, carelessness, blundering; possibly what counsel for defendant call an error of judgment, hut more accurately described as inadvertence, or a failure to properly notice 'or take into account. Had the claims been properly drawn at first, or after repeated rejection and objection, no abandonment of any kind would have occurred; hut the failure of the solicitors to know how to accomplish this was what caused abandonment. I cannot understand how so able an expert as Mr. See could say in his testimony that he could find no indication that these solicitors did not understand the invention. If they [562]*562did, their course is utterly inexcusable, in the view I take. Compare the eighth claim in the reissue with the twenty-fifth already quoted. This will illustrate the difference between the wrong way and the right, and serve to show why the solicitors were so pressed by the examiner that they abandoned all their claims for a computing scale:
“8. In a computing scale, the combination of a price-computing table, L. provided with graduations and numbers indicating the values of different quantities of commodities at different prices per pound, the disk or hub, D", provided with the knife-edge trunnions, D', projecting from its front and rear sides and resting in supporting-bearings, E, the index-hand, H, rigidly secured at its lower end to the disk, D'', and at its upper end co-operating with the table, L, and provided with a series of numbers indicating prices per pound, the vertically-depending weighted pendulum, D, rigidly secured at its upper end to the disk, D”, the link-pivot, N, carried by the disk, D", and normally standing at a point above the pivotal axis of the latter, means carried by the disk, I/', for counterbalancing the weight of the index-hand, H, and link-pivot, N, the scale-beam, A, fulcrumed near its outer end upon a suitable support at a and connected at its inner end with the link-pivot, N, by the link, E, the platform, R, mounted on the beam, A, between the fulcrum,. a, and the link, E, the tare-beam, A', mounted on the scale-beam, A, the poise, a', mounted on said tare-beam, the dash-pot, Z, and the piston, Z", moving therein, and having the upper end of its rod, Z', connected to the scale-beam adjacent the link, E ; substantially as and for the purpose set forth.”
It is a clear case of oversight or inadvertence.
De Vilbiss tried to get a patent upon a machine invented by him, but failed to secure one covering the real discovery. When 3,000 machines had .been sold, and it was discovered that the patent really afforded no protection, no intervening public rights having arisen, the grant was surrendered, and a reissue applied for, which would secure the real invention; the requisite grounds being alleged. There was no attempt to expand the invention, to obtain claims broader than the discovery, to patent the mode of operation, function, or result of the machine, to take in later discoveries, or to procure a different invention. If the patentee had sought to cover by the reissue the progress made by others in the same art, or attemptéd to narrow the rights of the public, or had not exercised reasonable diligence, the reissued patent could not properly be sustained. But under the rule of liberal construction applicable to cases of mere correction of mistake by Top-liff v. Topliff, and other cases cited above, I think the patent sued on should be sustained, if good on other grounds.
If the real invention had been claimed in- the original application, the claims rejected by the Commissioner, acquiesced in and abandoned by the patentee, could not properly have been included in the reissue. The rejection being intentional on the Commissioner’s part, and premeditated by the claimant, the relinquishment could not be the result of inadvertence, mistake, or accident. Mahn v. Harwood; Corbin, etc., Co. v. Eagle Lock Co. But the solicitors, never having claimed the real invention, did not abandon it by accepting counts sq narrow as to defeat it. Pressed by a technical and stubborn examiner, who seems to have been pursuing what has been called the “division fad,” but who had fair grounds of rejection by reason of the nature of the claims asked, the solicitors took the path of least resistance, and [563]*563accepted useless claims, which were the result of their initial incapability or inadvertence.
A further objection is the alleged express concession that the original patent lacked novelty and validity. I do not understand that any such concession was made. In the original specification, after describing the scale-beam, fulcrum, and platform support, the specification says that “nothing new is claimed lor this much of the device.” This simply means that novelty is not claimed for a scale-beam, fulcrum, platform, and support; not that a combination of these with 40 other elements is not new. If this language bears the construction claimed for it, the evidence of the solicitors’ incompetence would be conclusive.
Upon the issue of patentability the question is whether there was a good combination, or merely an aggregation of old elements, still performing old functions, without a new result, or an improved result. The patented scale was the first successful pendulum computing scale, and the first scale of any kind to use a tare-beam successfully in combination with the main scale and platform. Uike most combinations of old elements, the advance is not great. This was admitted by complainant’s solicitors in taking out the reissue. But there was a decided forward step. While there is not a new mode of operation, except the feature of the dash-pot, nor a new result, yet the old result is attained in a more beneficial and efficient way. Old elements never before brought together in any machine co-operate to produce the old result in a modified and improved way. The elements act upon each other to produce the total price. It may be difficult to see that the old elements in their new position operate in a way different from the old wav: but there is a unitary structure in which the action of one element modifies the action of others so as to produce an improved result. Thus, the action of the pendulum and index-hand are modified or checked by the action of the dash-pot as never done before in this way in a computing scale. And the action of the whole structure is modified by the tare device in the only successful way ever accomplished, except in defendant’s machine. Their c.oaction is precisely as so well described by the Court of Appeals of the Third Circuit in National Cash Register Co. v. American Cash Register, 53 Fed. 367, 3 C. C. A. 559, approved in this circuit in Dayton Malleable Iron Co. v. Forster, 153 Fed. 201. See, also, in the Court of Appeals of this circuit, Deere v. Rock Island Plow Co., 84 Fed. 171, 28 C. C. A. 308. The combination is easily distinguished from the elevator-leg case (Dunbar v. Eastern El. Co., 81 Fed. 201, 26 C. C. A. 330), the rubber tip pencil case (20 Wall. 498, 22 L. Ed. 410), the Globe-Wernicke Case (119 Fed. 696, 56 C. C. A. 304), and the Waterman fountain pen case (Reckendorfer v. Fabere, 92 U. S. 347, 23 L. Ed. 719). It is significant that very few combinations of old devices have been held void for aggregation. The fact that this machine was the first to be commercially successful should also have some favorable influence on the matter of construction. The conclusion reached is thought to be supported by the settled rule of this circuit in the recent case of Kellogg Switchboard & Supply Co. v. International Tel. Co. (C. C.) 158 [564]*564Fed. 104; International Tel. Mfg. Co. v. Kellogg, etc., Co., 171 Fed. 651, 96 C. C. A. 395.
Fifty-two prior patents were pleaded and proved to show aggregation and want of novelty, 81 of which are for pendulum scales. None of them were on the market when De Vilbiss entered the freíd. It is, indeed, quite remarkable that the best specimen of pendulum scales in the prior art should be the first one, the Sampson invention of 1849. This is the one the defendant says it has copied in some respects. Such an example of inverted advance in an art is quite instructive. As Mr. Macomber says:
“The analogousness of an anticipation is ‘proportioned inversely to the distance from which it is brought.’ ”
Tike gravitation, it decreases according to the square of the distance. Certainly a patent which caused no advance in the art should have but little effect on a successful one of 50 years later. Daylight, etc., Co. v. American, etc., Co., 148 Fed. 454, 73 C. C. A. 570.
After the Toledo scale had become successful, and defendant felt the competition in its business of marketing spring computing scales, it adopted the Toledo (with the variations mentioned later) so closely as to external form that it is difficult to tell them apart. Instead of doing this, why did it not copy the Sampson scale? Why did not the art progress after Sampson had pointed out the way? The difficulty' of answering such questions as this is what induces the courts to sustain successful combinations containing the elements of unsuccessful earlier patents. Strange as it may appear, the Sampson patent, now seeming to contain fully the general principle of both the Toledo and Moneyweight machines, had no influence on the art for 50 years. The Smyth patent of 1855 and the Guseman device of 1859 are clear retro-gressions. The same is true of the Take patent of 1866, McCormick of 1870, McKeone of the same year, Parr of 1874, Greenwood of 1877, Follett of 1878, Weaver of 1879, Atwater of 1888, Ball of 1888, Sir-rett of 1884, Keeler of 1884, Cartwright of 1888, Wolner of 1888, Knight of 1888, Conen of 1891, Chase & Smock of 1898, and De Vil-biss & Pool of 1899. Not one of these bears any resemblance to the Toledo machine. Apparently none had any commercial success. The record is replete with all the devices of the Toledo machine, and scores of others; but most of the scales themselves are awkward and complicated. Outside the dash-pot and tare device, nothing seems to have been discovered since the Sampson invention. The art was practically at a standstill when the Toledo scales appeared. They are simple, compact, and attractive in appearance, as their prompt adoption by defendant, even to the color of the paint, sufficiently shows. As said in Forsyth v. Garlock, 148 Fed. 461, 73 C. C. A. 577, the citation of so many patents in defense tends to strengthen plaintiff’s position, rather than weaken it. Possibly the Jensen patent of 1879, No. 881,-836, may be said to advance the pendulum scale art, but it is materially different from the Toledo, covering no main beam or lever, tare device, or dash-pot. I think the Toledo invention is meritorious and valid, and should be sustained.
[565]*565
[566]*566The question of infringement is less difficult than either that of reissue ■ or anticipation. The two scales are here shown without the housing, and it will be seen that the chief difference is in substituting the cam, D, for the disk, D, and the strap, F, for the link, F. The only-practical difference is that the strap has a longer and more uniform arc-action than the link; but the only practical resultáis that the chart must be changed correspondingly, a matter wholly mechanical. There are also other slight differences, chiefly that of hanging the pendulum obliquely instead of vertically, a thing without the slightest importance, the shape of the scale-beam, and a double instead of single tare-beam. Mr. See is manifestly mistaken in stating that the two pendulums have a different operation. Defendant adds nothing to the Toledo scale, certainly nothing which destroys the unity of the former, or makes it something else; nor does it leave out anything. It is true, as defendant says, that it copied very closely the prior art, for it copied the Toledo scale. But as this is a perversion of the true meaning of its statement, it is a complete answer to this suggestion to say that only form was changed, not substance, and in a case where form is not of the essence of the combination. Nothing is added which changes its function; nothing is left out which does so.- Several elements are changed, especially the link, link-pivot, disk, and disk bearings; but the changed devices do the same work as .those for which they are substituted, so that infringement is not avoided. The alterations are merely formal equivalents, performing similar functions, within the rule of Gould v. Rees, 15 Wall. 187, 21 L. Ed. 39; Gill v. Wells, 22 Wall. 1, 22 L. Ed. 699; Cochran v. Deener, 94 U. S. 780, 24 L. Ed. 139; Cantrell v. Wallick, 117 U. S. 689, 6 Sup. Ct. 970, 29 L. Ed. 1017; and National Cash Register Co. v. Cash Indicator Co., 156 U. S. 502, 15 Sup. Ct. 434, 39 L. Ed. 511.
It is insisted that damages should be increased on the ground that the infringement was malicious and conducted in bad faith. The record shows the following: Plaintiff paid $190,000 for the De Vilbiss patents and business, of which $142,500 was in cash. No other pendulum scale was then on the market; but defendant was then marketing a platform computing scale, and a spring balance computing scale. The latter was chiefly sold to meat markets, and the computing table was so arranged as to give the butcher the advantage of each odd fraction, thus cheating the customers enough to save the hire of a clerk, in a large market, being about 3 per cent, of the sales. See 142 Fed. 919, 74 C. C. A. 89. In 1903, a year and a half after the application for reissue was filed, but before the patent issued, defendant commenced to market a scale like the Toledo, but not enough like it in appearance to be confused with it, and later it changed to its present form of scale, copying the general style and form of plaintiff’s.scale, covering all its elements or equivalents, even painting it the same color, and calling it “The Genuine,” and the Toledo “The Imitation.” Large numbers of circulars containing pictures of the two scales, labeled as indicated, were sent out by defendant from 1902 to 1905. It also claimed in its advertising matter that the Toledo scale infringed the Phelps patent of 1890, owned by it, for a platform .computing scale, [567]*567and it afterwards brought suit for infringement of this patent against complainant, but did not prosecute it. The two scales bear little resemblance to each other. The price of the Toledo was $75, and of defendant's scale at first $55, and later $J-0 and $19. The latest form of defendant’s scale, and the method of competition, greatly injured the sale of the Toledo. The price was increased from $49 to $90, but a $49 scale, slight!}1' altered in construction, was still kept on hand, to be sold only as a last resort to meet competition. The record is full of the details of the most vicious and reckless forms of competition practiced by defendant, both before and after the reissue. Down to 1905 circulars were sent out claiming that the Toledo was an infringement of the Phelps patent, and four infringement suits were brought, one against plaintiff and three against users or sellers. The first was not brought to hearing, and the others were dismissed for want of prosecution. The claim of infringing the Phelps patent seems quite far-fetched, and indicates bad faitli on defendant’s part, in connection with the failure to prosecute the suits.
The fact that defendant’s unfair methods were commenced before the reissue, when there was no question of validity of reissue, aggregation, or anticipation between the parties, while showing an aggravated case of unfair competition, should not subject defendant to increased damages prior to the reissue, because it was not then knowingly infringing. After the reissue, the validity and novelty were debatable questions; and though the unfair competition was continued, even more unfairly than before, defendant undoubtedly considered the reissue an invalid one, which it might treat with contempt. While it seems that increased damages from the date of the reissue might be given under the Topliff Case, where the validity of the reissue must have been considered quite doubtful in view of prior decisions .of the Supreme Court, yet 1 am inclined to follow the rules stated by Judge Coxe, in Fox v. Knickerbocker Eng. Co., 165 Fed. 442, 91 C. C. A. 386, and Brown Bag, etc., Co. v. Drohen (C. C. A.) 175 Fed. 576, disallowing increased damages.
A decree is directed according to the prayer of the bill, except as to increased damages.