2 F.2d 655 (1924)
SANDY MacGREGOR CO. et al.
VACO GRIP CO.
Circuit Court of Appeals, Sixth Circuit.
December 11, 1924.
Hadley F. Freeman, of Cleveland, Ohio (Smith & Freeman, of Cleveland, Ohio, on the brief), for appellants.
F. E. Shannon, of Akron, Ohio, for appellee.
Before DENISON, DONAHUE, and KNAPPEN, Circuit Judges.
DENISON, Circuit Judge.
Infringement suit on patents, 1,439,338 and 1,439,339, both issued to Harry W. Smith on December 19, 1922, for a "golf practicing and exercising device." Application for 1,439,338 was filed December 17, 1920, and for the other one on June 1, 1921. Of the claims sued upon, the District Court held some invalid and other valid and infringed, and directed the usual injunction. The defendant appeals; the plaintiff does not.
In the immediate field of golf practicing devices there were several instances of a box set down so that its surface was flush with the ground, and carrying a vertical post projecting above the top so as to be like an ordinary tee. These posts, at the upper end, carried a permanently attached ball. At the lower end they were hinged, and in front of them the top of the box was cut away so that a blow upon the ball would tip the post over and carry the ball beneath the surface. A spring would return it to its upright position; or the standard itself was sometimes flexible enough to bend under the blow and then automatically return. It was also old to have balls or projectiles of different kinds connected by an elastic cord to a base to which they would return after being thrown or driven away; and this idea had been applied, in a general way, to golf practice balls. Smith's device consisted essentially of a mat and a ball and an elastic cord permanently connecting the two. There was nothing patentable in this basic combination. A sufficient elastic cord on top of the mat would be in the way; he therefore put it under the mat, letting the end which was to be attached to the ball go up through a hole near the front of the mat. The weight of the mat or the player would bind this cord to the ground; so he put it in a horizontal groove in the under side of the mat running backward from the hole. The mat end of the cord he fastened to a button depending from the mat at the rear end of the groove. The recessed groove or the depending button, or both, prevented any binding contact between cord, mat, and ground. An ordinary round ball drawn back to its starting point by an elastic cord might not quickly steady itself upon its seat; he therefore attached permanently to the ball at the place where the cord was fastened a flat bottomed cylindrical extension through which the cord passed axially, and which served as a tee when the ball was in place, and went and came with the ball, and aided in a quick and rapid seating in its return. These were the features of the first patent. By the improvements of the second patent he provided that the elastic cord should consist of a hollow rubber inflated tube; that it should be so attached to the ball that the latter could rotate thereon; and that, instead of having a hole entirely within the mat area, he might locate it in the front edge of the mat in the form of a notch, using the apex thereof as his guide for the cord. By the last expedient he avoided the friction of the cord going up through the hole and danger of shearing it off by the blow of the club. Each of the claims sustained in each of the patents is limited to the presence of one or more of these details, in addition to the basic mat-cord-ball combination.
We have little or no hesitation in finding invention in these patents. At first glance the matter seems very simple; and the District Judge at the outset doubted whether the large sales which were shown indicated "the utility of the device or the senility of the purchasers"; but he became convinced, as we are, that there was utility, both commercial and patentable. There was a prompt development of substantial sales, which seemed to be growing rapidly. Whatever the hints of the prior art, it does not seem that any device had ever come into actual use which permitted practicing with anything like the efficiency and success shown by this. The defendants, familiar with the patentee's developing commercial success, and not otherwise brought to the subject, copied the devices as closely as they thought they could without infringement, and persist in their desire to use them all. This is well sufficient to show utility, without regard to the principle that such a defendant will not be heard to deny it. Faultless Co. v. Star Co. (C. C. A. 6) 202 F. 927, 930, 121 C. C. A. 285.
Sufficient practical utility to make a device "useful," in the sense of the patent statute, is, we suppose, what is meant by the cases which say that an infringing defendant is estopped to deny utility. The phrase "patentable utility" is sometimes perhaps rather inaccurately used to denote the function of being useful plus the skill of the inventor as distinguished from that of the mechanic. Of course, the infringer is not estopped to deny patentable utility in this sense, because if he were he could never dispute patentability. When some courts, and this among them, have, in the course of denying patentability, thrown doubts upon utility, doubtless they have intended, as we have, to refer to the lack of any useful function or new result serving to give inventive character to the slight step which the patentee had taken, in differentiation from the prior art. See, for example, Naceskid Co. v. Purdue (C. C. A. 6, decided October 7, 1924) 1 F.(2d) 924.
The second patent is attacked because the details therein claimed are said not to show any invention as compared with the disclosure of the earlier numbered patent; and this earlier numbered patent is said to be, with reference to the other, a part of the prior art.
We can see no basis for this latter proposition. It loses sight of the fact that the question of prior art comes into the patent law through the statutory provision that a patent shall be invalid if the patentee was not the first inventor. The whole question obviously must have reference, not to the date of the patent, but to the date of the patentee's invention.
Nor do we understand how, save for some arbitrary purpose and by arbitrary rule, there can be any priority of grant or of publication as between two patents issued to the same patentee on the same day. Courts ordinarily decline to recognize fractions of a day; but to find this priority between simultaneous patents bearing successive numbers distinctions of time would come down to a matter of seconds. The issue clerk might and probably would recognize the request of the patentee as to priority in number between two patents on the same day; but, lacking such request and compliance, the result would depend upon which one the clerk happened to pick up first out of the several hundred issuing on that day.
It seems to us to be beyond dispute, upon the principles involved, that, if the combination a b c when first made by the patentee was invention, and if the addition of the element d adds utility, even though of itself it would not involve invention if compared with the earlier invention of a b c by some one else, yet the patentee is entitled to claims upon a b c and upon a b c d, and the validity of the second claim may rest, in part, upon its inclusion of the invention more broadly stated in a b c. Most naturally these claims would appear as generic and specific in the same patent; but, if the rules of the Patent Office require, or if the patentee desires and the rules of the Patent Office permit, the issue of separate patents to the same inventor, and they are issued the same day, it cannot be said that the one which bears the earlier application date or issue number is part of the prior art as against the other one — always assuming that "prior art" is a matter not touching dedication or double patenting. This was the conclusion we reached upon a discussion of the authorities in Higgin Co. v. Watson, 263 F. 378, 385, and even in a case where the patents were not simultaneous. The order of their issue was not important in that case save as to the question of double patenting. In the present case the patentee, Smith, at the date of filing his second application, might have canceled the first one and filed an application showing both the original and improved forms of his device, and containing most, if not all, of the claims here in suit. That he did not do so but kept his improvement patent separate, at the same time avoiding any question of dedication or double patenting, ought not to deprive him of the benefit of otherwise valid claims.
It is urged upon us that Underwood v. Gerber, 149 U. S. 224, 13 S. Ct. 854, 37 L. Ed. 710, is to the contrary. We do not so interpret it, though, without direct reference to or any consideration of the fact that the patents were simultaneous, Mr. Justice Blatchford uses language, some of which indicates that he regarded what was shown in the first numbered patent as necessarily "existing" before the invention of the other patent. In that case two patents were applied for on the same day by the same inventor upon identical specifications, and were later issued on the same day to the same patentee. One was for a composition of matter; the other was for the same composition spread thinly upon paper (like carbon paper). The patent for the composition received the earlier of the two successive numbers. For some undisclosed reason, the patentee sued only upon the later numbered patent, that for the coated paper, and, after the bill had been dismissed in the trial court, the plaintiff patentee refused an offered opportunity to amend its bill and sue upon both patents. The natural thought would be that plaintiff knew the composition of matter was old, and that the patent thereon was therefore invalid. The Supreme Court held the patent sued upon to be invalid. It held that there was no invention in taking an existing combination which was suitable to spread upon paper and had no utility except for that purpose and making the actual application to the paper. So far as the result was based (if it was) on the thought that by its disclosure in the first numbered patent the composition had become earlier art as to the other patent, we venture to think (as Judge Putnam did in Simonds v. Hathron [C. C.] 90 F. 210, 211) that this was without due regard to the fact that the background of earlier art against which the question of invention by the patentee must be judged is the background existing at the date of his invention, and not at his later patent date. Upon the whole, we do not think Underwood v. Gerber should be taken as deciding this proposition. Beyond the statutory defenses against the validity of a patent, the Supreme Court has found it necessary to imply another, viz. that there could be but one patent for one invention, and thus it has developed the defense of double patenting. The most persuasive reason for this rule of invalidity was that otherwise the patentee could extend his monopoly beyond the statutory period; but in Underwood v. Gerber it was thought that the splitting up of one indivisible right into two and subjecting the infringer to suits by two different owners of the right infringed justified applying the defense of double patenting as against two patents issued on the same day. Of course, the substance must be existing before it could be spread on paper. The case was practically one of inseparable process and product. The composition was not a useful or completed invention until it was spread upon the paper; the specification of the composition patent described the spreading; and the claim by necessary implication included the spreading. We take this to be the principle of decision which should be deduced from Underwood v. Gerber rather than any inference from the unnecessary reference to "an existing substance."
We therefore conclude that the presence of, for example, the inflated rubber tube as the extension member in the second numbered patent instead of the rubber cord of the first numbered, though it might not by itself be invention if some one else had earlier invented the first numbered, yet when claimed in connection with and by way of improvement upon the first form, it may serve to differentiate a valid claim in the improvement patent, as if the two were one. While it has been said that each claim is a separate patent, it has never been thought that a specified claim which does not disclose patentable invention over a more generic one in the same patent was, for that reason alone, invalid. When the claims are in separate patents, possible divergent ownership suggests complications, but we do not foresee any which compel hesitation in our conclusion.
Having found that there is substantial invention in the mat-slot-hole-cord-ball combination, there is no reason for limiting any one of these features to the precise form shown, nor for not applying a fair range of equivalents. As to those narrower claims, concerning which the contention of noninfringement is more forceful, the subject is not important. The broader claims being sustained, the narrower ones cannot affect the injunction. Accounting was waived.
The decree is affirmed.
 The case has been cited only twice by the Supreme Court. The first was in a case under consideration at the same time (Brigham v. Coffin, 149 U. S. 557, 562, 13 S. Ct. 939, 37 L. Ed. 845), and the citation was only as to the character of the supposed inventive step. The other was at the next term, 1894, in Miller v. Eagle Co., 151 U. S. 186, 198, 14 S. Ct. 310, 38 L. Ed. 121, and does not directly indicate the force now claimed. It has several times been cited by the lower courts to the effect that the issue of a patent was the dedication of everything not claimed, or that an integral invention cannot be divided, but, so far as we find, never necessarily to the effect that the earlier numbered of two simultaneous patents was prior art against the other one, save by Judge Lacombe at the Circuit, who seems, without discussion, to have accepted the case as authority to that point. Writing Co. v. Elliott Co. (C. C.) 106 F. 507, 508. In Doig v. Morgan Co. (C. C. A. 2) 122 F. 460, 462, 59 C. C. A. 616, there is a citation, but the earlier patent there involved was issued May 1, 1883, and the patent in suit May 18, 1886, on application filed August, 1884. (See Official Gazette.) In Benjamin Co. v. Dale Co. (C. C. A. 2) 158 F. 617, 619, 85 C. C. A. 439, the case is cited as holding that "the patent numbered first takes precedence of the others," but the court held the two patents to be entitled to stand as one. If, by the Benjamin-Dale Case and those there cited, the Second Circuit Court of Appeals accepts the theory that the first numbered of two simultaneous patents to the same person is prior art as affecting the presence of invention in the other patent, we feel compelled to differ.