This is a bill in equity, founded upon the alleged infringement of letters patent No. 228,186, dated May 25, 1880, and letters patent No. 231,199, dated August 17, 1880, for improvements in machines for making buckle levers. Buckle levers are the part of a buckle which is so made as to hold the strap by friction, being a substitute for the ordinary buckle tongue, which penetrates a hole in the strap. They are extensively used upon “arctic” rubber shoes. The two patents are for different parts of the same complex machine.
It is not denied that the defendant has infringed the first, third, and fourth claims of No. 228,136, and the second, third, fourth, and fifth claims of No. 231,199. Infringement of the fifth claim of the earlier patent, and of the first claim of the later patent, is denied. Although, as is apparent from the defendant’s admission in regard to infringement, the two machines are very similar, they differ somewhat in form.
In the plaintiff’s machine, after the two slots have been punched in the blank, and it has been bent into a U shape, and has been thrust upon the end of a mandrel, a pair of movable dies “are advanced, one on either side of the mandrel and blank, on lines substantially at right angles to the fiat sides of the U shape and of the mandrel. These dies have their engaging faces — that is, the faces that engage with the U-shaped blank — of such form and contour that when acting *722upon the U-shaped blank in combination with the mandrel, they bend or swage it into” a partially-formed lever.
In the defendant’s machine, after the TJ-shaped blank had been pushed upon the end of the mandrel, “there were two dies, — one on either side of the mandrel opposite to this U-shaped blank, — one of the dies — the lower one — being stationary in close proximity to the mandrel, the other die — the upper one — reciprocating vertically; the opposing faces of these dies having a configuration the reverse of that part of the mandrel occupied by the U-shaped blank, and having also portions of their surfaces adapted to form the bit of the buckle lever. As soon as the U-shaped blank was placed on the mandrel the upper movable die descended, and by means of this die, the lower stationary die, and the mandrel, the U-shaped blank” was bent into the same partially-formed lever which was produced by the corresponding operations of the plaintiff’s machine. In the plaintiff’s machine the sides of the mandrel and the faces of the dies are vertical, while in the defendant’s machine the mandrel is horizontal, with a stationary die below it and a vertically-reciprocatiijg die above it.
The fifth claim of patent No. 228,136 is as follows:
“In amachine for making buckle,levers the combination with the mandrel, m, and dies, n, n1, of the springs, n2, n2, to press the dies forward into proper position relative to the mandrel, substantially as set forth.”
In the plaintiff’s machine there are two things connected with the dies “to press the dies forward into proper position relative to the mandrel” when the dies are not swaging the metal. The object of this pressure is said in the specification to be “in order that a slight looseness of the operating parts shall not permit the dies to be so far thrust backward as to permit the buckle levers to telescope or overlap each'other as they push each other forward upon the mandrel.”
In the defendant’s machine, as used when this action was commenced, there was one spring “interposed between a bed plate and an arm of the mandrel support to hold the mandrel in close contact with the lower stationary die, and to press the mandrel and blanks downward towards or upon the lower die,” for the purpose of preventing the blanks from telescoping as they are being thrust along upon the mandrel. This spring has been removed, and, it is said, without injury to the efficiency of the machine; but it is plain that the object of introducing the spring was to prevent telescoping. The substitution of one spring for two, and the location upon the mandrel instead *723of upon the dies, are immaterial variations of the form of this part of the mechanism.
The first claim of No. 231,199 is as follows:
“ In a machino lor making bucklo levers, the combination of the mandrel, m, provided with the ribs, m, ms, of the dies, n, w-, o, o, and a support which presses the part, u, of the lever against the rib, in, substantially as set forth.”
In the specification the support is described as follows:
“ It will be understood that the wedging tongue, n2, supports the lower end of the part, u, of the lever firmly, while the dies, n, n1, are compressing the metal upon the mandrel; but owing to the wedge-shape of this tongue, when the die, n, is withdrawn from the mandrel the blank is released from upward pressure against the mandrel, so that it can bo readily fed forward to the dies, o, o.”
In the defendant’s machine the support is not fixed to the movable die, as in the plaintiffs mechanism, nor is it yielding. It is attached to the fixed die and performs the office which is described in the first claim. The defendant’s expert thinks that its machine does not infringe this claim, because the mandrel is not provided with the rib, m3, and because the second pair of dies, o, o, does not exist, and because the support is not removed after the action of the dies. The rib, m3, in the defendant’s machine is on the fixed die, which is the stationary part of that machine, while the mandrel is the stationary member of the plaintiff’s forming mechanism. The ribs have the same office in the two machines. The dies, o, o, were formerly on the plaintiff’s machine, but have been removed since this suit was commenced. So fa,r as is disclosed in the testimony, the patented machine was the first one to make “bended” arctics in one operation. The patent was not confined to the mere form of mechanism which the inventor adopted, and the “support” is not limited by the first claim to one which moves away after the action of the dies.
The main defence is that the two patents, Nos. 228,186 and 231,-199, with the exception of the fifth claim of the former and the first claim of the latter, are void, because the machine, with the exception of the mechanism described in said claims, was in public use for more than two years before the date of the application for letters patent. But one application, dated December 2, 1878, for a patent upon the entire machine was originally brought. This was afterwards divided into two applications upon different parts of tlie same machine. In order to defeat the patents, the machine must have been in public use before December 2, 1876. Graham v. McCormick, 21 O. G. 1533.
*724The facts are that from 1862 to 1868 the patentee made another kind of buckle from those produced by this machine upon two or more different machines. Between 1868 and the fall of 1878 another kind of buckle was made by one machine. For a year prior to the fall of 1874 he made the “beaded” buckles, i. e., the kind now under consideration, upon two machines. In 1874 he ordered the skeleton of the patented machine from Bliss & Williams, his workmen or himself making the patented portions. This machine was in a condition in which it was used to manufacture bucke levers in the fall of 1874, and continued to be so used, without substantial change, until the spring of 1878; but it was not a perfected invention. It had two defects, — one that it choked, and the overlapping blanks had to be picked apart by a workman; another that the bead was not parallel with the slot, because the blank could not be forced upon the mandrel evenly. Nevertheless, it was used, in some seclusion from the public, to make levers, and it made about 50,000 gross, which were sold; but the organization was defective until it was perfected in the early part of 1878 after repeated experiments. The inventor always adhered to the idea of perfecting the invention and then obtaining a patent upon it. The two improvements which were introduced in 1878 were the springs between the levers and the dies, which prevented overlapping, and the rib, m3, in order to keep the blank in position when it was forced upon the mandrel. These changes, which are apparently not of great importance, perfected the invention, and enabled the inventor to take the final step between partial and complete success.
It is perfectly true that a patentee cannot be permitted to use for profit a machine, which embodies a perfected invention, for a period of two years or more, and then obtain a valid patent for the old machine by means of the addition of some new improvements which, in the language of Judge Lowell, “were intended to benefit the patent rather than the machine.” The present case is that of a machine which was imperfect, and which demanded and received the continuous experiments of the inventor to remedy the defects in its organization. It is not true that the inventor cannot safely use for profit such a machine in its imperfect state, lest two years should elapse during the experimental period before the invention is completed and the patent is applied for.
The plaintiff is entitled 'to a decree for an injunction and for an accounting.