The former die, it will be understood, is the chamber into which the clay is forced by the expressing screw, and from which it emerges in the form of a column of stiff clay onto the propulsion belt, to be cut into brick lengths. The patentee says in the specifications of the former die that:

"The improvement consists in making the top and bottom of the former die convex, and its sides straight or more or less concave, * instead of.

*

*

as heretofore, making the sides of the same convex. The new form, I find by experience, is a great improvement upon the old ones, as the body of clay, which is retarded in the middle by the convexity of the former die at the top and bottom, is better spread out laterally, and the clay more forcibly packed in the corners, than was the case when the sides of the former die were convex."

It is not denied that the defendant uses just such a former die as that described and claimed in this patent, but it is contended that such a former die is shown in the prior art, and that it lacks novelty. Here is, as the learned judge at the circuit said, "another instance where a question of doubt is presented,-whether that which was discovered is to be regarded as in the nature of invention, or, on the other hand, of supplying by mere mechanical skill the remedy which the result of the operation of the machine suggested." In 1863 Chambers took out a patent for a former die having both the top and bottom and the sides convex. In explaining the defect he was attempting to remedy, he said:"

"It will be obvious upon reflection that the ordinary operation of a plunger or other propelling device in a machine with the ordinary form of die is to produce the greatest amount of velocity in the center of the mass, giving the outer edges and surfaces less, and hence rendering them more liable to be made ragged and broken by partial adhesion to the die while passing through it. The velocity being less, the density is also less of these outer portions; in other words, the quantity of matter in a given space is greater at the center than at the surface of the exuded bar of clay. The remedy for this is to be sought in a reversal of the ordinary disposition of the material, forcing the greatest amount of clay into the corners of the brick or tile, and compressing it there so that the last action of the die upon it will be to give it smoothness, instead of tearing it, and rendering it rough and ragged.. The peculiar form of my dies completely effects this object."

He then describes the die:

"The cross section of this die is at its inner end circular and at its outer end rectangular, as seen in Figure 4. A cross section on a line (midway between) is shown in Figure 2.

"In this figure we see the angles or corners rounded out or grooved, and these grooves gradually tapering till they disappear altogether at the angles. a, a, of the rectangular opening of the die. These grooves constitute the main peculiarity of the invention, their object and effect being to crowd a

greater quantity of clay into the angles of the bar of clay as it passes through the die, so as to give them greater solidity and firmness, in accordance with the views hereinbefore stated."

The only change which the patentee made in the die of patent No. 275,467 was to remove the convexity of the sides, and substitute either straight or concave sides. This was merely a modification of the earlier former in degree, and not in principle. In the earlier Kells patent a former die is shown with its sides convexed and its top and bottom straight. It is true that the convexity is carried into the die itself so as to make the bricks of concave sides. But the convexity of the sides is said by the patentee to be adopted for the purpose of securing sharp corners, a purpose quite similar to that of Chambers in the device under consideration. On the whole, we are constrained to deny validity to this claim.

The decree of the circuit court is affirmed as to claim No. 24 of patent No. 297,671, and is reversed as to claims No. 7, 9, 10, 11, and 12 of patent No. 362,204 on the ground of noninfringement; and as to claim No. 2 of patent No. 297,675, claim No. 6 of patent No. 207,343 and claim No. 1 of patent No. 275,467 on the ground that the claims are invalid for want of patentable invention; and the case is remanded to the circuit court, with directions to dismiss the bill as to all the claims here involved except No. 24 of patent No. 297,671. The costs of the appeal will be taxed one-fourth to the appellant and three-fourths to the appellee.

HART & HEGEMAN MFG. CO. v. ANCHOR ELECTRIC CO. et al. (Circuit Court of Appeals, First Circuit. March 13, 1899.)

1. PATENTS-REISSUES-VALIDITY.

No. 238.

Where a reissue is granted to correct an error of a single word in the specifications, as by changing "hole" to "slot," and a corresponding change is made in a single feature of one of the numerous figures in the drawing, but no change is made or needed in the claim, there is no reason for holding the reissue invalid.

2. SAME-INFRINGEMENT-ELECTRIC SWITCHES.

The only difference between a patented electric switch and an alleged infringing switch was that in the former the catch was released by a movement radially inward, while in the latter the release was by a movement radially outward, and the former was operated by a flat spring, one end of which was attached to a stud depending from a spring plate. while the latter was operated by a spiral spring, the corresponding end of which was attached either to a small cap at the top of the hub just beneath the operating handle, or was fastened by being cast through the hub itself, the cap in that case being omitted. Held, that these variations involved merely the use of mechanical equivalents, and the patent was infringed.

3. SAME.

The Hart reissue, No. 11,395 (original No. 459,706), for an electric snap switch, construed, and held not anticipated, valid, and infringed.

Appeal from the Circuit Court of the United States for the District of Massachusetts.

92 F.-42

This was a suit in equity by the Hart & Hegeman Manufacturing Company against the Anchor Electric Company and its officers for alleged infringement of a patent for an electric snap switch. The circuit court held that the patent must be construed narrowly, and consequently was not infringed. 82 Fed. 911. The complainant thereupon appealed to this court.

Charles E. Mitchell and Henry B. Brownell, for appellant.
Edward P. Payson, for appellees.

Before COLT, Circuit Judge, and BROWN and LOWELL, District Judges.

LOWELL, District Judge. This was a bill in equity brought for the infringement of reissued latters patent No. 11,395, granted to Gerald W. Hart, December 12, 1893, for electric snap switches. The alleged infringing switch was made in accordance with the description of letters patent No. 547,149, granted to Norman Marshall, October 1, 1895. To the complainant's patent the defendants have raised objections, both formal and substantial.

As the patent in suit shows two specific constructions, the defendants contend that the complainant must be limited to one of these. Both constructions, however, are embodiments of the same inventive idea, and both are sufficiently covered by the claim. Hart's original patent taken out for the invention in question was No. 459,706. The reissue was granted to correct the error of a single word in the specification (by changing "hole" to "slot"), and a corresponding modifica tion was made in a single feature of one of the 11 figures illustrating that patent. The error was caused by oversight, and was unconnected with the gist of the invention, though its adoption rendered the machine inoperative. No change was made or needed in the claim. If the patentee made a meritorious invention, he ought not to lose the benefit of it by reason of a defect so narrow and technical. See Walk. Pat. § 218a.

We pass to the substantial objections made by the defendants to the maintenance of this suit. The only claim of the reissue is as follows:

"The herein-described snap switch, consisting of a stop plate having stopping shoulders, a central hub, an operating handle, an eccentric moving with said hub, a switch plate, a spring plate, a spring, and a catch operated by said eccentric for releasing and stopping the switch plate, substantially as described, and for the purpose specified."

new.

It is obvious that none of the elements set forth in the claim are The invention, if any, must be found in the combination of these elements. Again, it was not new in the art to make electric snap switches. Plainly, none of those made prior to Hart's invention were so satisfactory in their operation as to check the demand for improvement, or the patenting of machines devised to secure it. An examination of some of the many patents and machines introduced in evidence in this case suggests the cause of the failure of the earlier devices, a failure sometimes complete and sometimes partial. The machines made under the patent in suit have gone into large general use, though there is no evidence that the plaintiff has monopolized the

manufacture of positively operating electric snap switches. Hart's combination of stop plate, hub with rigidly connected eccentric operating radially, spring, and catch, seems to us to contain important novelty. Certainly his invention was not primary, but we think that it was of value, unless anticipated in the prior art.

To show anticipation, the defendants have introduced in evidence many patents. We need mention but a few of these; for, if no one of those mentioned anticipated the complainant's invention, there was no anticipation in any of those passed over. The Bourne British patent, No. 15,617, resembles Hart's in little except the presence of a spring and eccentric. It has no stop plate, no locking against backward movement, no certainty of operation. Its switch plate is oper ated directly by the eccentric, and not indirectly, as are the switch plates of both complainant and defendants. It is a far cruder machine than Hart's or the defendants', though one of the same general class. Between it and Hart's we think there was meritorious and considerable invention; yet upon the anticipation alleged to be found in the Bourne device the defendants very largely rely.

The Norton patent, No. 430,252, contains no eccentric, strictly speaking, and its catch moves vertically, not radially. These differences, and others of less importance, give it a mode of operation quite unlike that of the patent in suit. No. 376,976, issued to Bergmann, is a reciprocating switch. Its operation is altogether different from Hart's, and its resemblance to the latter quite remote.

The Davis patent, No. 476,613, comes closer to the complainant's device. The complainant has introduced considerable direct evidence that his invention was made before that of Davis, though the latter was first applied for. This evidence was not shaken on cross-examination, and there is nothing to control it. It is true that evidence of prior invention, unsustained by proceedings to obtain a patent, is properly regarded with suspicion, but in this case we are inclined to find it sufficient. Even if, however, the Davis invention be taken to be prior to that of Hart, we think it does not anticipate Hart's switch. The Davis switch has no eccentric, properly so called, and its operation is materially different. It may be true that an imaginable combination of the Bourne and Davis patents would closely resemble the patent in suit, but such a combination is not obvious, and would require patentable invention.

Defendants rely also upon Hart's prior patent, No. 447,728. There is uncontrolled evidence that the switch described in the patent in suit was first invented and first reduced to practice. In any case, it was invented before the issue of No. 447,728. That patent, therefore, is not in the prior art, properly so called, and has priority only as a prior patent issued to the same inventor. We think that it is not so nearly identical with the patent in suit as to deprive the latter of real and useful novelty. The earlier Hart patent has no eccentric, properly so called, and no radially moving catch, and its stop plate, if there be one, is very different from that of the patent in suit. The entire operation of the mechanism is different. It may be added that the difference between some of the patents above mentioned, such as those of Bergmann and Davis, and the patent in suit, appears more

plainly when all are embodied in actual machines than when the specifications, drawings, and claims alone are examined.

It seems to us, then, that the patent in suit represents a valuable and useful, though limited, invention. If it be really valuable, we think there can be little doubt that it is infringed by the defendants' machine. The only differences between the two machines are: (1) The defendants' catch is released by a movement radially outward, while in the patent in suit the releasing movement is radially inward. In this respect, the two devices are plainly the mechanical equivalents, the one of the other. (2) The patent in suit is operated by a flat spring, one end of which is attached to a stud depending from a spring plate. The defendants' switch is operated by a spiral spring, the corresponding end of which is attached either to a small cap at the top of the hub and just beneath the operating handle, or is fastened by being passed through the hub itself, the cap in that case being omitted. That the spiral spring is the equivalent of the flat spring is clear, and it is equally clear that the defendants' cap in which the end of the spring is inserted is the mechanical equivalent of the complainant's spring plate and depending stud. It cannot help the defendants that in some of their machines the end of the spring is thrust through the hub itself instead of thrust into the cap. In considerable degree this is recognized, even by their expert, Mr. Freeman, who

testified:

"If the Hart claim covers any equivalent of the spring plate, then the number of elements in the defendants' switches would be the same as the number of elements of complainant's switch broadly stated in the claim of complainant's patent. So too, broadly stated, the general mode of operation of defendants' switches is substantially the same as the general mode of operation of complainant's switches."

The learned judge in the circuit court seems to have been of the same opinion, for he says:

"Moreover, if the court was able to ascertain that the complainant's device was of a broad character, indicating a substantial advance in the art, it might be justified in holding that, although the spring plate is omitted in the respondents' device, yet inasmuch as, taken as a whole, it has what is equivalent to the complainant's device as a whole, including the substance of it, the complainant's patent should, therefore, be construed liberally and broadly, so that any infringement might be prevented if found."

As our examination of the prior art has led us to the opinion that the complainant's device did indicate a substantial advance in the art, it follows that the Hart claim does cover a mechanical equivalent of its spring plate. The difference between the two switches seems to us to be merely that which usually distinguishes an infringing machine from that which it infringes. Nonpatentability, rather than noninfringement, is the substantial defense to the action. The defendant corporation was therefore liable.

Two minor questions arise in dealing with the case. The defendants' answer alleged that the complainant had made or sold his switches without marking them or the packages containing them "Patented," and without notifying the defendants of any alleged infringement. In Sessions v. Romadka, 145 U. S. 29, 49, 12 Sup. Ct. 799, 805, a similar objection was raised, but the court said: