This bill was brought for alleged infringement of letters patent No. 476,506, for “an improvement in wick-raiser attachments for lanterns,” granted on June 7, 1892, to Frederick K. Wright, assignor to the Steam Gauge & Lantern Company. The patent has two claims which are as follows:

“(1) Tlie combination, with a lantern frame and a detachable bottom, secured to the lantern frame by a releasable fastening, of an oil pot permanently attached to said bottom and capable of rotative movement therein, a burner provided with a laterally projecting wick-raiser shaft, and a stop-arm attached to the lantern frame and adapted to come in contact with the bottom of the wick-raiser shaft, substantially as set forth.

“(2) The combination, with the lantern frame and a detachable bottom, secured to the lantern frame by a releasable fastening, of an oil pot capable of rotative movement therein, and provided with beads or projections below and above the said bottom, a burner provided with a laterally projecting wick-raiser shaft, and a stop-arm attached to the lantern frame and adapted to come in contact with the bottom of the wick-raiser shaft, substantially as set forth.”

Several defenses were set up in the court below, including want of evidence to show title in Raymond, trustee, who sued, and lack of necessary parties plaintiff; but the court considered only the defense of noninfringement, which the court sustained, and thereupon dismissed the bill.

The invention of the patent in suit relates to a class of lanterns, theretofore well known, in which the wick is raised or lowered by a rotative movement of the oil pot carrying the wick-raiser shaft. The claims are for combinations, and one of the elements of each of the two claims is “a stop-arm attached to the lantern frame and adapted to come in contact with the button of the wick-raiser shaft.” This member of the combination of each claim is described in the specification thus:

“N represents an upright stop-arm, which is secured with its lower end to the inner side of the collar, A, and extends upwardly on the inner side of the globe to the vicinity of the button, m, where it terminates with a number of inwardly projecting teeth, n, which are adapted to come in contact with the button and turn the same when the oil pot is rotated within the lantern.”

This stop-arm device has a small number of teeth, forming a short segmental rack which extends over a comparatively short part of the circular path of the wick-raiser button ; and the toothed head of the stop device is supported by an arm which extends downwardly from the head, and is affixed at its lower end to the inner side of the body hoop or lower ring of the lantern frame. The defendants’ structure, instead of the short segmental rack, has a complete toothed ring or circular gear face with which the pinion of the wick raiser engages, and the toothed annulus or circular head is secured to the upper end of a cylindrical shell or casing surrounding the oil pot, which shell or casing at its lower end is affixed to the lan*868tern bottom. The above-noted differences in construction result in these differences in operation, namely, Wright’s segmental rack permits of engagement with the button of the wick-raiser shaft during an extremely short period of its rotation, whereas the defendants’ circular gear face secures the continuous engagement of the pinion therewith; and again, while Wright’s stop-arm remains in the lantern frame when the bottom and oil pot are removed, in the defendants’ structure when the bottom of the lantern is removed the wick raising and lowering mechanism is also removed and remains constantly in perfect operative position.

It is certain that Wright was not a pioneer inventor in lantern construction. The class of lanterns which he essayed to improve was old. The real basis of his improvement, namely, means for-raising and lowering the wick without removing the oil pot from the body of the lantern, did not originate with Wright. He found it already in the art. The patent to Merrill of 1865, No. 46,010, shows a rotatable oil cup, a fixed rack, and a stem provided with a pinion for engaging the rack, whereby the wick may be raised and lowered by a rotary movement of the oil cup without its removal. The same'method of adjusting the wick, by a wick raising and lowering attachment, is found in the patent to Feeny of 1889, No. 411,-218. Feeny’s wick-raising device in construction is substantially the same as Wright’s. In principle it is identical. Feeny shows, a rotatable oil pot, an arm-supported segmental bar provided with inwardly projecting teeth, and a wick-raising spindle carrying a toothed disk or pinion adapted to engage the teeth of the segmental bar, and the spindle is operated by the rotary movement of the oil pot. Indeed, the only difference between the structure of Feeny and Wright is that Feeny’s oil pot is removably connected with the body of the lantern, while Wright’s oil pot is permanently affixed to a detachable bottom. But Wright did not invent and was not the first to employ in connection with lantern construction a detachable bottom secured to the lantern frame by a releasing fastening. For this feature of construction we need only refer to the patent to Miller of 1875, No. 162,090.

The prior state of the art, we think, forbids any broad construction of the claims of the patent in suit. The rule here applicable, it seems to us, is that laid down in McCormick v. Talcott, 20 How.. 402, 405, 15 L. Ed. 930, where, distinguishing between a primary invention and a mere improvement of a device, the Supreme Court said:

“But if the invention claimed be itself but an improvement on a known machine by a mere change of form or combination of parts, the patentee cannot treat another as an infringer who has improved the original machine by use of a different form or combination performing the same functions. The inventor of the first improvement cannot invoke the doctrine of equivalents to suppress all other improvements which are not colorable invasions of the first.”

The witnesses all agree as to the structural differences we have noted above between the defendants’ wick raising and lowering-mechanism and Wright’s. But, as was to be expected, the expert *869witnesses on the one side and on the other disagree as to the materiality of these variations between the two structures. An attentive examination of the proofs, however, has brought us to the conclusion that the differences which distinguish the wick-adjusting device of the defendants from the wick-adjusting device of the Wright patent are not colorable but substantial, and that the defendants' construction is not an invasion of Wright's patented improvement. By reason of the defendants’ circular gear face being in continuous engagement with the teeth of the pinion, their wick-adjusting mechanism is always under the control of the operator, which is not the case in the device of the patent in suit. Again, while in Wright's structure, upon the removal of the lantern bottom from the frame, the toothed head of the stop-arm is disengaged from the toothed button of the wick-raiser shaft, the defendants’ rack and pinion remain in permanent engagement, so that when, after removal, the lantern bottom is replaced, their wick-raiser mechanism is operative at once without any adjustment of parts.

We may add that the evidence warrants the conclusion that Wright’s lantern never went into practical use, whereas the defendants’ lantern has gone into extensive public use. Of course, these facts as to use are by no means decisive, but they tend to negative the charge of infringement made in this bill.

Upon the whole case we are of opinion that the Circuit Court was right in holding upon the proofs that the defendants had not infringed the plaintiff’s patent, and accordingly the decree of the court dismissing the bill of complaint is affirmed.