after making the foregoing statement, delivered the opinion of the court.
It is objected at the threshold that this court is not at liberty on this appeal to inquire into the validity of the first claim of the Martin patent, because, before the hearing was had in the court below, the validity of that claim had been determined by the United States circuit court of appeals for the Eighth circuit in the case, which will here be called the “Dempster Case,” of Mast, Poos & Co. v. Dempster Mill Mfg. Co., 49 U. S. App. 508, 27 C. C. A. 191, and 82 Fed. 327. The court below, in obedience to the opinion of this court in Electric Mfg. Co. v. Edison Electric Light Co., 18 U. S. App. 641, 10 C. C. A. 106, and 61 Fed. 834, deemed itself bound to follow the earlier decision, and now it is insisted that this court must affirm the order entered without inquiry into the question of the patentable novelty of the claim or into any question'decided by the court in the Eighth circuit.
The decisions touching the practice on appeals from interlocutory orders, under the judiciary act of 1891, have not been in entire harmony; but in the recent case of Smith v. Vulcan Iron Works, 165 U. S. 518, 17 Sup. Ct. 407, where the decisions touching the subject are collected, the supreme court has defined clearly the scope of the review which the act was intended to authorize. After declaring that the appeal, which by section 7 of the act may be taken from an “interlocutory order or decree granting or continuing such injunction,” is an appeal “from the whole of such interlocutory order or decree, and not from that part of it only which grants or continues an injunction,” the court proceeds to say that the manifest intention of the provision was “not only to permit the defendant to obtain immediate relief from an injunction, the continuance of which throughout the progress of the cause might seriously affect his interest, but also to save both parties from the expense of further litigation, should the appellate court be of opinion that the plaintiff was not entitled to an injunction because his bill had no equity to support it.” The comprehensive terms of this expression forbid the suggestion that it does not apply when the appeal is from an order made upon affidavits, and not from a decree ordering both an injunction and an accounting, entered as the result *337of a hearing upon full proofs. If there is ground for a distinction in tha t respect, it is in favor of the appeal from a preliminary order made upon ex paite and imperfect showings at the commencement of litigation, rather than an appeal from an injunction perpetual in terms granted after a full hearing, which is called interlocutory only because there remains to be taken an accounting, upon which the evidence adduced cannot ordinarily affect the injunction. This being the scope of the appeal, the logical inference would seem to be that every application to a circuit court for an injunction or temporary restraining order should be considered on its merits, and that a ruling or opinion of another court upon any question involved should be given only its just a ud reasonable weight according to the circumstances. The statute gives the light of appeal; the supreme court has determined that the review, so far as may be, shall extend to the merits; and it is not consistent to say that the decision of an inferior court must be pronounced on one basis and reviewed on another.
In respect to the merits: It was stated in the opinion of the court in the Dempster Case, and we think correctly, that “the essential element” of the first claim of the patent was “the internal toothed spur wheel or spur gear meshing with and driving the pinion which actuated the pitman and pump rod”; and at the same time it was conceded to be “true that internal toothed spur wheels, their effect, and their relative advantages over external toothed wheels had been familiar to mechanics time out of mind.” In view of that concession, and of the obvious cogency of the dissenting opinion of Judge Thayer, it is to be inferred that the court could not have found in the patent the novelty essential to invention but: for the presence of features of evidence not apparent in this record. For instance, in that case only one patent on windmills, Ao. 182,394, granted on September 19, 3870, to Edward Williams, was offered as anticipating Martin’s, and that showed a pitman actuated by two eccentric external toothed gear wheels. Accordingly the court, after rejecting as unsatisfactory the testimony of the one witness by whom it was attempted to show prior use. dismissed from further consideration the defenses of prior use and nonutility, and, in considering the question of infringement, declared, upon the proof before it, that, prior to Martin’s invention, “all windmills had been driven by external toothed spur wheels”; that in the mills so constructed, as the cogs and other parts wore away, a pounding and racking of the machinery was caused “as the pitman connection passed over the center, and the motion changed from a pulling to a pushing one, and vice versa, shortening the life of the mill, and sometimes stripping the cogs from the pinion”; and, after quoting from the specification the statement that “a plurality of the pinion teeth are always engaged with the internal spur gear, resulting in giving a perfectly uniform and smooth and noiseless reciprocating motion to the actuating rod, thereby prolonging the life of the machine by saving it from constant jarring and preventing wear and tear,” the court, added: “The evidence is undisputed that this invention completely accomplished its purpose.” Certainly a remarkable success, as stated, and well worthy of a patent. But the proof in the present record shows no such merit. In that case the record *338also disclosed, and the court laid stress upon the fact, that the president of the infringing company, before abandoning the external and adopting the internal gearing, had seen one of the mills made under the Martin patent in operation, and had appropriated that form of construction on the advice of his pattern maker, who had been in the employ of an earlier infringer. “The inevitable conclusion to which these facts lead,” the court declared, “could not be escaped by asking that the broad terms of the first claim be confined to the meaning of the restricted terms of the other claims”; and the contention that the substitution of internal for external gearing in a windmill, in view of the common use of both forms of gearing in other machines in familiar use, was not invention, the court answered by,declaring it passing strange, if naught but the skill of the mechanic was required to make the improvement, “that no mechanic ever made it until after Martin discovered and • described it.” “Moreover,” the court added, “the combination of Martin immediately went into general use. More than three thousand windmills which contain his combination have been manufactured and sold since 1890.” That opinion was handed down on August 2, 1897, and the suggestion is obvious that the sale of the number of mills stated in a period of more than seven years was not phenomenally large for a strong and energetic establishment. The proof in this case shows that the prior form of construction was by no means superseded by the new, which, according to the evidence here presented, constitutes not more than 15 per cent, of the total output on the market. It is evident, too, that the supposed defects of the external gearing had been exaggerated; that their tendency to friction and to wear and tear, if constructed with adequate strength and properly adjusted, is not essentially different from that of the other form; and, if cogs have ever been stripped off, it was due to faulty construction or maladjustment. This seems to us to be the reasonable view of that phase of the question; and instead of it being-true here, as in the Dempster Case, that the alleged infringer appropriated Martin’s construction after seeing it in operation, the testimony of the president of the Stover Company is that the adoption by that company of the internal gear for operating the actuating rod of its windmill was a mere accident or incident of construction in the first instance, without any knowledge of the Martin patent or mechanism, and was accomplished simply by transferring the internal gear from other machines made by the company side by side with its windmills, such as power feed mills driven by horse power. Ao reason is shown to discredit this testimony, and it follows that, if Martin made an invention, the appellant is entitled to the credit of making the same discovery. The more reasonable conclusion is, as it seems to us, that neither can be credited with the exercise of more than ordinary mechanical skill. That the internal gearing employed to actuate the pitman of a windmill has some advantages, and therefore may be properly called an improvement over external gearings,- there can be no doubt, but in a windmill the advantages are not different in kind from what they are in other machines which have a dead point, the passing of which involves a constant change from pulling to pushing and vice versa in the application of the operative force. The advan*339fagos are the same “relative advantages” which, it is conceded, had been familiar to mechanics time out of mind. That the improvement was not incorporated in windmills sooner, on the proof here made, is not strange. It is shown, and without proof would be inferred, that it is a matter of expense and inconvenience to remodel and recast, any part of a machine, though in itself trivial; and upon this ground, as well as by reason of that inertia which disposes manufacturers to let well enough alone, a manifestly desirable change in the structure of a machine, already in successful use, will go unmade until forced in order to meet the competition of a new or more enterprising rival. The first to make such an improvement is not in fact an inventor, and to infer invention merely because it had not been produced sooner would be to reach a false conclusion by an argument neither persuasive nor plausible. See Parlin & Orendorff Co. v. Moline Plow Co., 89 Fed. 329.
Other considerations aside, the proof of the prior art in this record is different from and far more extensive than it was in the Dempster (’ase. Desides the Williams patent, a number of patents are in evidence showing windmills with both external and internal gearings. The fact was pointed out and deemed material by the court below that the internal gear shown in each of the patents in evidence “had no function relating to the driving of tin; shaft, or the creation of reciprocal action”; the function in nearly all the cases being, it was said, “merely to keep the wheel in the wind.” In the Eighth circuit the court, accepting it as true that prior to Martiir's invention all windmills had been driven by external gearings, held that, though the like use of such gearings in other machine's had long been familiar, the adaptation or transfer thereof to the windmill evinced invention,- and the court below, on proof that such gearing liad been in common use in windmills, but not in immediate conned ion with the driving-shaft or to product' reciprocating motion, was constrained, as already explained, to hold that the difference between the eases was not such as to justify a different conclusion. It may be that the presence of the internal gearing in other parts of a windmill ought to be regarded as of little or no more significance' in respect to the particular combination covered by Martin's claim than such gearing in familiar use in machines used for sawing wood, mowing, and oilier purposes, especially since in some of the latter live use was to produce reciprocating motion; hut in view of the fact that in the windmill of Perkins and in numerous others, on which patents had been granted before the date of Martin’s application, external and internal gearings were'in common use side by side, we find it impossible to credit with invent Lon a, patentee who did nothing but put one of the forms in the place of the other in a machine where both were already present, producing by the change no new result, but only improved effects or relative advantage's long known to mechanics to be the result, of using that form instead of the other. Such an exchange or substituí ion of one of two forms of gearing for another in a machine, where, both are in familiar use side by side, is no more to be called invention than would be the substitution, under like circumstances, of one familiar make of pulley for ano I her.
*340The customary reference has been made to such cases as Crane v. Price, Webst. Pat. Cas. 393, 409; Electric Co. v. La Rue, 139 U. S. 601, 11 Sup. Ct. 670; and Potts & Co. v. Creager, 155 U. S. 597, 15 Sup. Ct. 194, — but the distinction is clear. In Crane v. Price the discovery was of the susceptibility of anthracite to the action of the hot blast. 1 Rob. Pat. § 266, note. Martin discovered no new quality in internal gear wheels, and applied them to no use which can be properly called new. In Electric Co. v. La Rue the invention related to telegraphic keys, and consisted “in substituting for the trunnions or pivots upon which the lever vibrates a torsional spring or strip of metal”; the spring not only taking the place of the pivots or trunnions, but, when used in connection with certain adjusting screws, taking the place of the ordinary retractile spring. The court, after stating that there was nothing in the exhibits in the case to show “the use of a torsional spring in a telegraphic instrument,” and that the invention “did not seem to be one of great importance,” said: “We think the adaptation of this somewhat unfamiliar spring to this new use, and its consequent simplification of mechanism, justly entitles the patentee to the rights of an inventor.” There was there not simply the substitution of one known form of spring for another. The retractile spring was displaced by the torsional spring, and, if there had been nothing more, it would have been essentially like the putting of the internal gear wheel in the place of the external; but the replacement of a fixed and inflexible pivot or trunnion with a torsional spring was an essentially different achievement, rising, as the supreme court considered, to the dignity of invention. But that conclusion, it is to be observed, was reached on the grounds that the torsional spring was somewhat unfamiliar, had never before been employed in a telegraphic instrument, and, as introduced, subserved a new use. Here, the contrary is true in every particular. The internal gearing, its effect, and its relative advantages over the other form had been familiar. It had been in use in windmills side by side with the external wheel, and if, as employed in the Martin combination, it served a use which, in any sense, was new, it was, in the language of the opinion in Potts & Co. v. Creager, “so nearly analogous to the former one that the applicability of the device to its new use would occur to a person of ordinary mechanical skill.”
It is not perceived that further proofs are possible of a character to change the result. The decree or order below is therefore reversed, with directions to dismiss the bill for want of equity.