(PiNney and Newman, JJ., concurring). "With the decision that the judgment of the circuit court must be reversed and a new trial granted, all concur, but on the question of whether there was sufficient evidence to warrant the court in submitting the case to the jury differ-, enees exist of such a character as to render it necessary that a separate opinion be filed on one branch of the case.
The evidence is undisputed that there was no discoverable defect in the construction of the machine, or in its repair, at the time of the accident. It was of the same construction as machines ordinarily used and found safe for the business for which it was designed. The boy testified that it made an unexpected revolution the day before he was injured, without any pressure being put upon the pedal; that it started from a state of rest without any such pressure, and revolved once, at the time of the accident; and one witness testified to seeing it subsequently do the same thing when no one was near it. But the evidence of those who worked with the machine for a long time prior to the time the boy commenced working1, and who worked with it for a lone1 time *284after the injury, wTas to the effect that the machine had always worked perfectly. The testimony was undisputed that the mechanical arrangement was such that its motions were necessarily absolute and positive, depending wholly on pressure upon and release of the pedal. From this state of tlm record, giving to the plaintiff’s evidence the most favorable construction it will reasonably bear, including all reasonable inferences therefrom, would it warrant a jury in finding that the machine, though free from all discoverable defects, was out of repair, nevertheless, and that the defendant ought to have known of it and to have apprehended that a personal injury to the plaintiff might probably result therefrom 1 That is the question.
In our opinion,.the question stated in the preceding paragraph must be answered in the negative; that it will not, consistent with settled principles, admit of any other answer; and that, when the authorities bearing on the subject are carefully considered, no conflict will be found, either in the adjudications of this court or els.ewhere.
There are many cases that hold that an unexplained accident with a machine, not liable to occur if such machine was properly constructed and in a proper state of repair, is evidence of negligence; as in Spaulding v. C. & N. W. R. Co. 30 Wis. 110, where it was held that the escape of fire from a passing locomotive engine, sufficient to cause damage, raised a presumption of improper construction or insufficient repair or negligent handling of such engine. To the same effect are Cummings v. Nat. Furnace Co. 60 Wis. 603; Kurz & Huttenlocker Ice Co. v. M. & M. R. Co. 84 Wis. 171; Stacy v. M., L. S. & W. R. Co. 85 Wis. 225; Mullen v. St. John, 57 N. Y. 567; Volkmar v. Manhattan R. Co. 134 N. Y. 418; McCarragher v. Rogers, 120 N. Y. 526, and many others that might be cited. Such cases lay down a very well recognized principle in the law of negligence, 'but do not reach the question here under discussion, and do not conflict in the slightest degree with *285numerous authorities that go on another principle, just as well recognized and firmly established, to the effect that undisputed prooí of freedom of the machine from all discoverable defects, either in construction or repair, effectually overcomes any mere inference or presumption arising from the happening of the accident, so as to leave no question in that regard for the jury; as in Spaulding v. C. & N. W. R. Co. 33 Wis. 582, where this court held the inference that a locomotive engine was defective, arising merely from the escape of fire therefrom sufficient to cause damage, rebutted by conclusive proof that the engine was free from discoverable defects, so as to leave nothing on that point for the consideration of a jury. Also, to the same effect, are Read v. Morse, 34 Wis. 315, and Gibbons v. W. V. R. Co. 62 Wis. 546.
The decisions along this line in this court, reaching from an early day to the present, are believed to be without conflict, and, if like harmony does not exist elsewhere,— which we do not admit,— a decision of the question at this time should not be of a nature to create doubt respecting the rule that obtains here. In the recent case of Menominee River S. & D. Co. v. M. & N. R. Co. 91 Wis. 447, this court, in an exhaustive opinion by' Mr. Justice Pestney, went over the whole ground, collating and commenting on substantially all the important adjudications in this court, and reaffirmed without exception the doctrine that, where negligence arises as a presumption of fact from the happening of an accident with machinery,— not a mere inference to be drawn by the jury, but a presumption so convincing and persuasive that, ■unexplained or unrebutted, it prima facie establishes the alleged negligence,— such presumption is, nevertheless, overcome by conclusive proof that the machine was free from discoverable defects, so as to take the question in that regard from the jury. To the same effect is the still more recent ■case of Badger v. Janesville Cotton Mills, 95 Wis. 599, where an operative, who was standing on a ladder endeavoring to *286remove a belt from á rapidly revolving shaft, was caught by the arm, carried around the shaft, and severely injured. He claimed that the ladder -was defective, that it broke by his weight upon it, and that thereby he was caused to make a movement which resulted in his arm being caught, with the result mentioned. There was positive proof that the ladder was free from all discoverable defects, and on that the trial court nonsuited the plaintiff, and this court sustained the ruling upon the ground, among others, that in the face of positive proof that the ladder was free from any defect that any one could point out, the evidence of plaintiff that it broke by his weight upon, it was not sufficient to send the question to a jury.
If time would permit, what has preceded, could be reinforced by numerous other citations from courts of the highest respectability; but, aside from Menominee River S. & D. Co. v. M. & N. R. Co., supra, probably no stronger or more satisfactory case can be found upon which to rest the decision of this court than Dingley v. Star Knitting Co. 134 N. Y. 552, for in its leading facts it is in all respects like the case under consideration. The injured person was a young boy. The machine was a breaker or card in a cotton mill, operated by a belt from a pulley on the main shaft to a tight pulley on the breaker. At the side of the tight pulley was a loose pulley. When the card was idle and the main shaft in motion, the belt was on the loose pulley; otherwise the .belt was on the tight pulley. The method of starting the card was to press the belt, by hand, from the loose pulley to the tight pulley, and of stopping it, to press the belt back to its former position. On the day in question the boy claimed that he pressed the belt to its place on the loose pulley, and stopped the card; that he then reached under the machine to do some cleaning, as was his duty to do; and that while so doing the machine unexpectedly started, and he was seterely injured. The boy testified, as *287in this case, that the machine bad operated that way once or twice before, and there was evidence on the part of plaintiff,, as here, that on some other occasion the same thing had occurred. The proof showed that there were no discoverable-defects; that the machine was the same as usually used and found safe against such accidents; and that it was operated a long time before and a long time after the occurrence in question without the happening of any such accident. The court held that, even if it were true that the machine had started unexpectedly, there was no warrant for saying that the defendant was negligent in respect to keeping it in repair, in the face of the undisputed evidence in respect to there being no defects that any one could point out; that to-Hold otherwise would be to extend the liability of the master beyond any bounds known to the law, and without any reason to support it. It is significant that the court further held, in effect, that the starting of the machine unexpectedly, after the belt was properly shifted to the loose pulley, was an occurrence so improbable that the evidence of the boy that it did so start not only did not warrant a finding in plaintiff’s favor on that point, but the legitimate inference from the whole case was that the belt was not properly shifted, and that neglect in that regard caused the mischief, and explained it all. That is in line with Cummings v. Nat. Furnace Co. 60 Wis. 603, where it was held, in effect, that proof that the machine was free from defects left the happening of the accident evidence of improper handling; and with Badger v. Janesville Cotton Mills, 95 Wis. 599, where it was held that proof that the ladder was free from defects left it so improbable that it broke from plaintiff’s weight upon it as to effectually overcoine his evidence that it did so break, and that where the probabilities from unmistakable physical facts are all one way, a jury has no right to-say the truth is the other, because there is evidence from the mouth of a witness to that effect, especially where such witness is *288a-n interested party. See, also, to the same point, Groesbeck v. C., M. & St. P. R. Co. 93 Wis. 505; Lens v. Whitcomb, post, p. 310; Payne v. C., R. I. & P. R. Co. 39 Iowa, 523; Artz v. C., R. I. & P. R. Co. 34 Iowa, 153; Bloomfield v. B. & W. R. Co. 74 Iowa, 607.
Further, on the main question we call attention to Redmond v. Delta L. Co. 96 Mich. 545, as directly in point. There a person was injured by an unexpected failure to operate of an endless log chain, which was controlled by pressing upon a lever to start, and releasing the lever to stop, it, as occasion required. On the occasion in question plaintiff claimed that the chain failed to respond to a release of the lever. The proof was that no defect existed in the apparatus that could be discovered or pointed out, and that the machine responded promptly to the manipulation of the lever before and after the accident. Counsel for plaintiff contended that the jury were warranted in finding that the failure of the machine to operate as usual at the time of the injury was owing to some defect therein. The court held .to the contrary, and that a finding on counsel’s contention ■would be based on mere surmise and conjecture that there was negligence somewhere; that if that were permissible, ■some of the fundamental principles of the law of negligence ■would be done away with; that where an employee seeks compensation from the master for injuries received in the latter’s employment, he must trace it to some fault of the master,— to some distinct failure of duty. See, also, Louisville & N. R. Co. v. Binion, 98 Ala. 570; Davidson v. Davidson 46 Minn. 117; Brymer v. S. P. R. Co. 90 Cal. 496; Morton v. D., B. C. & A. R. Co. 81 Mich. 423.
The preceding are not in conflict with McCarragher v. Rogers, supra, as expressly stated by the judge who wrote the opinion in Dingley v. Star Knitting Co., supra; nor are they with Volkmar v. Manhattan R. Co., supra, where the negligence charged was failure to keep a railway track in *289repair. The question of negligence was held to be for the jury for want of conclusive proof to overcome the inference of negligence arising from the physical facts. Neither is Dingley v. Star Knitting Co., supra, nor the principle here applied, in conflict with Connors v. Durite Mfg. Co. 156 Mass. 163, where the accident was caused from the unexpected movement of an engine caused by a leaky valve, or with Martineau v. Nat. Blank Book Co. 166 Mass. 4, where a machine started by the unexpected slipping of a belt from á ‘loose to a tight pulley, as in the Dingley Case, the difficulty being traced by evidence to a defective belt and a defective pulley, as the producing cause.
The foregoing sufficiently shows the substantial harmony ■existing to the effect that while, in a class of cases mentioned, an accident may be of such a character as to evidence negligence, and the inference in that regard may be so strong as to amount to a presumption of fact, not open to consideration by a jury unless explained or rebutted so as to’ throw doubt on the question, yet, ’when such inference or presumption arises from physical facts alone and points to insufficiency in the construction or repair of machinery ■as the producing cause of the accident, conclusive proof that such machinery was free from all discoverable defects will wholly overcome it, and leave no legitimate basis for a finding of defectiveness which the owner ought to have discovered, and reasonably to have apprehended a personal injury might probably occur to some person whose personal ■safety ordinary care required him to guard.
The principle here discussed we deem definitely established, as are substantially all principles pertaining to the law of negligence, the same as in any óther branch of the law, civil or criminal. Such principles are not, as is sometimes supposed, largely indefinite, indefinable, and easily bent to suit the exigencies of particular cases. Cases do ¡not make the law. Whether principles of law apply to a *290particular case depends upon the facts of such case, but that is no more true in the law of negligence than in the law of criminal, homicide, or any other branch of jurisprudence.
From the foregoing, yhile fully concurring that the judgment of the circuit court should be reversed for all the grounds stated in the opinion of the chief justice, we think that there was not sufficient evidence of negligence respecting the suitableness of the machine to take the case to the jury on that point.