This action for patent infringement is met with the customary defenses, invalidity and lack of infringement. On July 15, 1941, Patent No. 2,249,086, applied for October 22, 1940, was granted to Charles L. Macintosh for an expansible X-type lazy tongs bracelet having detachable links. This patent is now owned by plaintiff. The sole novelty was the method of detach-ability of the links. Only two prior patents have been cited having detachable links. In one, Surrows No. 2,172-525 — a non-expansible bracelet — the detachable device contained an additional locking plate. It teaches nothing of present moment. In the other, Magerhans No. 447,692, detachability was effected by removable screws. I am satisfied that, due to the necessarily small size of bracelet links, Magerhans was difficult, or at least expensive, to manufacture, and impractical for the primary purpose to which the detachability feature was addressed. A bracelet for watches, the principal use for expansible bracelets, must be a close fit, and if without removable links must be carried in three, or possibly five standard sizes. Since there are many models the amount of stock required presents a problem, especially for small retailers, of both space and expense. The most desirable bracelet is one that can be adjusted readily by a salesman at the moment of the sale, so that several sizes need not be carried. Facility in the rapid handling of jeweler-size screws in order to add or subtract links is not to be expected of all salesmen, and even a jeweler doubtless would not like the bother. That Magerhans did not meet the requirement of the trade is apparent from the fact that while this patent issued in 1891, plaintiff’s sales manager, with *290nearly 20 years experience, never saw or heard of one in production. Yet when plaintiff introduced detachable link bracelets they met shortly with great success.
The entire issue, however, is not as simple as the above recitation might indicate. It is to be noted that Macintosh obtained his patent in 1941. He manufactured under it, with indifferent results, from 1939 until 1947. I find that the principal reason for his then discontinuing was that the trade was so accustomed to carrying several sizes of bracelets that the saving introduced by only one small producer was scarcely worth the trouble of accustoming salesmen to adjusting the links. Consequently Macintosh, though more expensive to manufacture, could not command any higher price, and it was only when plaintiff, by means of a substantial promotional program, educated the trade, that success was possible. But there was another matter. Occasionally Macintosh detached accidentally in the hands of the wearer. The links are fastened together by headed vertical studs attached to one member reposing in the narrow portion of keyhole openings in a complementary member, the force of a spring, present inside each link to contract the bracelet, holding the studs in that position. Angular, or diagonal, location of this narrow portion permits intentional disengagement. In order to disconnect a link the bracelet is expanded to its full length and then a transverse movement will force the studs into the larger openings. Thereupon a lateral movement following the axis of the studs, accompanied usually by a slight distortion, will result in the heads of the studs passing through the openings effecting the disconnection. This process can be accomplished readily by a person of little experience. While the detaching position is one not normally occupied by the bracelet, plaintiff found, as Macintosh had, that it could occur, and the bracelet come apart accidentally. Before putting the device on the market plaintiff accordingly redesigned slightly the shape and location of the keyhole opening. This required an additional motion before disengaging, and eliminated any danger of accidental separation. Under date of October 4, 1954, plaintiff applied for a patent embodying this additional feature, which was issued as Burkhardt No. 2,765,615 on October 9, 1956.1
The commercial production of Macintosh in its basic state for some seven years demonstrates its practicality. I have no difficulty in finding invention. The ready detachability without tools due to the angular location of keyholes and the holding of the studs in place by the action of the springs while serving their other, and principal purpose, was a substantial novelty, and a useful one.
Infringement is another matter. The accused device is in effect Boots Patent No. 2,775,862 applied for December 28, 1955 and issued January 1, 1957. Boots, who is defendant’s president, testified that he first became interested in the subject of detachable links in May, 1954, when his salesmen showed him a foreign-made affair. Shortly thereafter he investigated Macintosh, but decided not to adopt it. His asserted reason was that he felt Macintosh would come apart, but I find this was not due to the so-*291called Macintosh defect, but a fundamental, and erroneous, objection to all keyhole devices as lacking in strength. Furthermore, his statement that May, 1954 was the date of his first interest in detachable links was untrue, because in 1925 and 1928 he had obtained patents on bracelets with detachable links. His present interest, however, stems from plaintiff’s success, and it was only after observing this that he recommenced designing.
I find, if material, that Boots designed his Patent No. 2,775,862 for the express purpose of circumventing Macintosh and plaintiff’s adaption. It is interesting to observe that that patent was not cited. There is no presumption that Boots does not infringe Macintosh. Herman v. Youngstown Car Mfg. Co., 6 Cir., 191 F. 579; Zachos v. Sherwin-Williams Co., 5 Cir., 164 F.2d 234. Other than the fact that not quite so much metal is stamped out of the member into which the studs go — a paper advantage, so far as the evidence indicates —Boots is substantially identical to Macintosh except that for the larger circular openings through which the heads of the studs pass there is substituted a slot on an inclination of some 10%. This change of plane makes the actual opening a flattened ellipse instead of a circle, but in a way this is simply a matter of appearance. Because the opening is inclined to the lateral plane of the bracelet, only part of the conceptual circle intersects the existing metal. There still would be essentially a circular opening if the metal were thicker.
Furthermore, it does not seem to me that a circle per se is important. What is important is the existence at the end of the angular positional slot of a larger opening through which the studs can pass. If defendant had duplicated Macintosh exactly, except to substitute square-headed studs and corresponding square openings, infringement would be obvious. However, what may be important, also, is that by inclining the larger opening out of plane Boots made necessary an additional motion to disengage the links. Boots cannot be separated unless the plane of the bracelet is bent backwards about 10% to permit the heads of the studs to pass through the slots. Such lateral distortion will practically never occur in actual use, so this effectively prevents accidental disengagement.2
Defendant further points out that not only does Boots require inclination, but it does not rely on the springs to hold it in position. This is merely a duplication of its claim, another way of saying that Boots will not separate without distortion. The springs are still useful, as they normally keep the heads of the studs away from that position where an unintentional movement would cause separation. But even eliminating this effect of the springs would be of no beneficial consequence, as the springs have to be present anyway.
In view of the substantial contribution made by Macintosh, and the essential similarity of Boots, I am disposed to regard Boots’ device as at least an equivalent unless what it accomplished was a marked improvement. Of course an improvement which is a straight addition to a patent does not cease to be an infringement simply because it is an improvement. Temco Elec. Motor Co. v. Apco Mfg. Co., 275 U.S. 319, 48 S.Ct. 170, 72 L.Ed. 298. It may be argued that Boots was not all ad*292dition, but in some measure substitution. In such event it would be appropriate to judge equivalency by the extent of the improvement — the significance of the departure in relation to the remaining basic concept. In spite of the recitations to be found in the Burkhardt wrapper, I do not believe the possible defect in Macintosh was that important.
Under the circumstances I do not find it necessary to make a final decision on the somewhat rarefied arguments of whether Boots was' a literal infringement, or a substitution as distinguished from a straight addition. I find Macintosh valid and infringed.
. The principal evidence as to the frequency with which the pure Macintosh would accidentally come apart came from the Burkhardt application and file wrapper. There is no estoppel, cf. Temco Elec. Motor Co. v. Apco Mfg. Co., 275 U.S. 319, 48 S.Ct. 170, 72 L.Ed. 298, but since plaintiff owned the invention throughout this period these statements must be treated as admissions. I am inclined to regard them, however, as considerably exaggerated. It is to be noted that Macintosh, Jr., in testifying why he and his father ceased manufacturing after some seven years, gave as his reason expense, and mentioned “some” complaints of accidental separation only as an afterthought. I believe if such separation had been a serious problem they would have discontinued production much sooner.
. It is true that the specifications in Macintosh provide that “a slight lateral movement or distortion of the linkage * * * will cause separation * * However, the word “or” is significant. Distortion, that is, inclination, in practice is desirable, because it is easier to center the expanded head of the stud in the larger opening of the keyhole by slightly distorting the lateral plane. But if the stud becomes precisely centered in the larger opening separation can be effected by a vertical (lateral) movement without distortion. This possibility is what resulted in accidental separation. Boots’ inclined slot avoided this defect by requiring lateral distortion. Prior to Boots, however, Burkhardt had already modified Macintosh (in another manner) to require distortion.