Plaintiff and appellant, a motorman, was operating, for the defendant and respondent, an electric car at a terminal where the car was turned around by means of a loop. A man of at least ordinary intelligence, he received the instruction usually given by defendant to its trainmen and had been in its service for more than a year. The car ran upon the curve of the lodp at full speed and was derailed and capsized. Thereby plaintiff received the injuries here complained of.
Plaintiff’s contention was as follows: He was seated upon his stool, and was grasping with his left hand the handle of the controller; ris hand being in contact partially with the brass and partially with the vood. Plis right hand was grasping the metal handle of the brake. His right foot was resting upon the dog of the hand brake, which may :>r may not have been in contact with the ratchet. He had upon his Iiands a pair of cotton gloves, being very much soiled, and the palms fceing impregnated with dirt, oil, and sweat. He wore cotton stockings ind leather shoes. His current was nearly at the maximum, if not quite, lie received a shock of electricity which froze his hands to the handles. By reason of the temporary paralysis of his arms he. was deprived ■f control of his car and prevented from shutting, off the current and ■pplying the brakes so as to reduce the speed. Defendant was charged Bdth negligence in failing to properly furnish and maintain reason-. Bbly safe appliances and instrumentalities with which to perform his Bmrk, and in failing to caution and warn the plaintiff of the hazards *506to which the performance of his duties exposed him and which were unknown to him.
The defendant contends that it was not negligent in these respects,, and that plaintiff did not in fact receive the shock of electricity at or immediately prior to the accident, but was asleep at his post,, and negligently ran his car at full speed upon the curve.
The controller was the apparatus which governed the current from the trolley wire overhead used in operating the car. It consisted of a central, movable part, known as a “barrel,” upon which were little copper blades, called “conductors,” which, as the barrel revolved, made contact with stationary “fingers” and sent the current through the parts in various combinations. It was surrounded by a casing or frame made of sheet iron. This was lined with a nonconducting material — asbestos—three-eighths of an inch thick. ' For purposes of this appeal it will be assumed that this metal controller casing or frame was liable to become charged with electricity; that such charge of the controller casing endangered the motorman, un less it was kept sufficiently grounded; that if the grounding of thel controller casing was sufficiently interrupted the motorman could re-| ceive a shock in the manner testified to by plaintiff at the trial. The| function of the ground wire was (1) to act as a means of escape-as a waste pipe, as it were — for discharging into the earth the sur plus or waste current from the interior of the controller, and als (2) to discharge into the earth any current which might have leakecj into the frame of the controller. A further device for grounding! put in by the defendant company, was a metallic brace attached to thcj back of the controller and passed to the metallic air brake contralle-frame, to which it was attached. It was fastened by a steel screv to the casing. If for any reason the contact with the frame by th strap became inadequate to ground the surplus electric current, danl ger would not be avoided by the fact that it was subsequently at tached to the brake. The case was twice tried in the district cour At the first trial the jury disagreed. On the second trial, at the clos| of all the testimony, the court directed a verdict for the defendan The propriety of that order is the question on this appeal.
1. The plaintiff himself testified to the fact that his hands weii frozen to the handles and as to the places on his hands and feet whet *507he had been burned by the current. The gloves he wore were produced in court. His testimony and that of other witnesses further tended to show that he had obeyed signals to stop for passengers at points so near the place of injury as to tend to negative the defendant’s contention that he was asleep. It is true that plaintiff's testimony as to the stopping was controverted by witnesses for the defendant. If this were all there were to the case, the issue should clearly have been submitted to the jury.
2. Plaintiff also introduced expert evidence which, he argues, showed the insufficiency of the grounding of the casing. Two experiments were made on the car, which had been brought in front of the trial courthouse for the purpose. Of these, the first was the “bell test.” The expert placed one wire from a dry battery with a voltage of one or one and a half on the controller; placed another wire at different times on the brass part of the air brake, not the handle, on the handle of the air brake, on the hand brake staff, on its support, and on the dog used to set the hand brake. The current was communicated, but the bell connected by a wire to the frame did not ring. The second experiment was a “magneto” test. It was made by means of a small dynamo discharging a current of “anywheres between three hundred and five hundred” volts. When the connection was made between the points stated, and the current turned on, “the bell barely tinkled.” These experiments were adduced to show that there was not sufficient “contact” between the casing and the brake frame through the metallic strap or brace previously described. Plaintiff’s own experts so testified. They testified, also, that it was thus proved that the ground wire designed to ground the controller casing was “either absent or a very poor one.”
For many reasons, some of which only will be presently set forth, we are at a loss to comprehend what strength this testimony adds to plaintiff’s case. It is true the controller was in the same condition at the time of the beginning of plaintiff’s experiments as it was at the first trial and as at the time of the accident. Upon critical examination of the record, it appears that^the admission of counsel for plaintiff that this was the case was not clear, although the trial court evidently regarded it as sufficient. If it be disregarded, we are of the I opinion that unimpeached testimony affirmatively showed such to be *508■the fact. None the less the experiments were not performed upon the car in the condition it was at the time of the accident. For plaintiff’s expert, finding the strap screwed up “fairly tight,” or so that • he “could loosen it with a small wrench,” testified that before the first electrical test was made “I loosened the bolt probably a turn and a half or two turns, and as the strap didn’t come off the casing, and the washer stuck out, I left it that way, and screwed the bolt back, and then subjected that first test. * * * Q. You made no test of this car 980 as it came there to Wabasha street until you had unbolted— A. Until I had loosened the bolt.” Therefore the experiments were inconclusive as to the strap or brace. ■
Moreover, the general conclusions of the experts were weak and uncertain. One of them was asked: “Now, will you state once more, in your own way, what the result of your experiments or tests showed with reference to the grounding of that casing at that time? A. Well, as far as I could see, the only indication was that the strap partially grounded the casing.” One expert testified the bell test “did show a slight connection. Q. Enough to be of a practical use or benefit in .the protection of the motoneer?. A. I couldn’t say. Q. You couldn’t¡ tell about that? A. Not at that time; no.” So, afterwards, when asked if the condition he found by his test was of such a character1 or condition as to render the happening of an injury as plaintiff described “reasonably possible,” an expert answered: “I should say| the possibility existed.” Finally, it is to be noted that at the time ofj the experiments the car, which, we repeat, had been brought down] by its own power to the tracks in front of the courthouse for these| experiments, was in the condition of ordinary operation.
The fundamental weakness in plaintiff’s test is that the ground wire| screwed into the casing had not been disconnected. One part of it: admitted function was to ground the electricity which might hav* leaked into the casing. That it was insufficient to accomplish thi was the basis of the negligence charged. If it availed to ground th* current in the casing, the motoneer in plaintiff’s position was admit-l tedly safe. The natural inference from plaintiff’s experiments isj that, when the current from either the dry battery or the magnet* apparatus was connected with the casing, the current tended to pas instantly to the earth through the ground wire. If the metallic strajj *509was loosened, the fact that the bell did not ring tended to show that this wire had performed its function and grounded the current. When the strap was tightened, apparently, the current divided, and a small part passed through the brace to the earth, and the balance through the ground wire. We are unable to avoid the conclusion that plaintiff’s own experiment tended to show that the means for grounding in fact provided were not negligent.
The analogy to the Morse telegraph system suggested by plaintiff fails to diminish the force of.this reasoning. If the sending instrument start a message over a wire supported by insulated poles, it would operate the receiving instrument and pass thence through the ground wire to the earth. If the poles should not be insulated, and [be good- conductors, the current would tend to pass through them to [the ground, and either no, of a diminished, current would go to the receiver. If it should receive no message, the indication would be that the current had been grounded. Here the controller, itself exclusive of grounding devices, corresponds to one telegraph pole not Insulated, the brake frame (speaking generally), itself exclusive of grounding devices, to the other, also not insulated, the battery or Inagneto apparatus to the sending instrument, the brace or strap to ftart of the wire, the wire from the frame to the bell to the rest of she wire, and the bell to the receiving instrument. The current sent Ito the controller would tend to go into the earth through its ground lure. If the brace should be disconnected from the casing, it would ■e like a parted wire, and in the absence of an arc through the air ■not here involved) naturally would not cause the bell to ring. If Bonnected from the casing, and the current communicated, it did not Bng the bell, the inference would be that the ground wire had fully Bmctionated. If the bell should ring but faintly, it would indicate B divided current, part passing through the ground wire and part Brough the brace.
■ On the other hand, defendant’s experts performed four different Bperiments, which were in their nature calculated to elicit the truth, Bd are subject to no reasonable criticism, except that they were per-Brmed on the day after the plaintiff’s, and after the plaintiff’s ex-[flrt had altered the connection between the brace and the casing. De-IBidant’s tests were made between the top of the casing and the track *510and wheels of the car. It is to be noted that, unlike defendant’s experiments, the plaintiff’s tests were not between the controller itself and the ground. Moreover, defendant used a current from the trolley wire with a current of about six hundred volts, while the plaintiff’s current was of less voltage. Defendant’s expert testimony was positively to the effect that the occurrence of a shock to the' plaintiff in the way in which he contended to have received it was impossible, and that his position was “absolutely safe.” The two classes of experiments, however, appear to have been largely consistent with each other, and to have indicated a proper grounding of the casing by means of a ground wire. The alteration by plaintiff in the strap in connection with the casing tends to deprive both of significance as to the ability of the strap and brace to ground the current. Taking all the experiments together, they incline to weaken, and not to strengthen] plaintiff’s case.
3. Plaintiff’s case may - properly have been regarded by the tri; court as resting substantially on plaintiff’s own testimony and on th doctrine of res ipsa loquitur. Defendant insists, however, that thi| maxim does not apply as between master and servant. As a univers. principle, this is not the law in this and a number of other state;] Ordinarily the application of the principle does not depend upon th relation between the parties, except indirectly, so far as that relatio] defines the measure of duty imposed on the defendant. Griffen v. Manice, 166 N. Y. 188, 59 N. E. 925, 52 L. R. A. 922, 82 Am. S. 630. There have been cases in which the servant has successful invoked it against his master. There is more force to defendant! further contention that the doctrine should not be applied where, here, the dangerous agency is under the immediate physical contri of, and in actual use by, an experienced servant, who complains of tJ injury. In the case at bar, however, it is not necessary to determij this controversy, nor to inquire into the exact nature of the so-call presumption, for the prima facie case of negligence arising from t] application of the maxim is rebuttable. From its application t court or jury may infer negligence, but neither is bound to do Carmody v. Boston, 162 Mass. 539, at page 542, 39 N. E. 184. T| presumption of negligence in current phrase, whose precision is here material, is purely one of fact, and may not survive suffici(| *511proof of due care by the person sought to be charged. It does not necessarily follow the plaintiff through the case, so as to compel the submission of the question of fact to the jury, as, for example, does the presumption of innocence accused in a criminal trial.
4. The trial court, giving to the plaintiff the benefit of the presumption, held that it was rebutted by the testimony as a whole, and that, taking the testimony as a whole, plaintiff had not borne the burden of proof resting upon him of showing negligence on the part of the defendant. In this view, we think, he must be affirmed. In addition to the inference of the proper condition of the controller, and especially of its grounding, from the experiments which have been described, and from the absence of perforation of the asbestos lining, and from other significant circumstances not necessary to be detailed here, the case is controlled primarily by two facts — the use of the car, and the facts attending its purchase and inspection. The car had been operated daily from November 8 to the time of the accident2 by various motormen, including the plaintiff, and reported “O. K.” each time returned to the station. On the day of the accident three motormen operated it without trouble before .it was turned over to plaintiff at 5:35 p. m. He then made several uneventful trips between that time and the time of the accident, about midnight. It has been operated since the accident without trouble of any kind. It was brought to the scene of the experiments by the safe operation of this identical controller. The continual use of the car in the condition it was from the time of the accident to the time of the trial without accident is a demonstration of the practical sufficiency of the grounding of the controller.
No testimony was adduced to show that the car was old or outworn. It appears, however, to have been used for some time. Its controller and the grounding of the controller were naturally subject to deterioration. The controller was of standard type and made by a reputable manufacturer. It was shown to have been inspected frequently, searchingly, and in a reasonable, practicable, and usual Inanner. To have visually examined the ground wire inclosed in insulating material would have involved the dismantling of the machine. The case is within the principle that “the law does not con*512template that railroad companies will in general make their own cars or engines, and they purchase them in the market, of persons supposed to be competent dealers, just as they buy their other articles. All that they can reasonably be expected to do is to purchase such cars and other necessaries as the3r have reason to believe will be safe and proper, giving them such inspection as is usual and practicable as they buy them. When they make such an examination, and discover no defects, they do all that is practicable, and it is no neglect to omit attempting what is impracticable. They have a right to assume that a dealer of good repute has also used such care as was incumbent upon him, and that the articles purchased of him which seem right are right in fact. Any other rule would make them liable for what is not negligence, and put them practically on the footing of insurers.” Campbell, C. J., in Grand Rapids v. Huntley, 38 Mich. 537, 547, 31 Am. 321. And see Carlson v. Phoenix, 132 N. Y. 273, 30 N. E. 750; Flood v. Western Union, 131 N. Y. 603, 30 N. E. 196 ; Reynolds v. Merchants’ Woolen Co., 168 Mass. 501, 47 N. E. 406 ; Westinghouse Electric & Mnfg. Co. v. Heimlich, 127 Fed. 92, 62 C. C. A. 92. That a company is not responsible when añ accident] occurs, because of a defect “which might have been discovered by a minute- examination, but which was not discovered in fact by such] an examination as was customary and reasonably practicable,” seel Richardson v. Great Eastern, R. R. 1 C. P. Div. 342 (Court of Appeals). The company was required to inspect with adequate careJ but not to dismantle complicated machinery for purposes of inspection. Clyde v. Richmond & D. R. Co. (C. C.) 65 Fed. 482, In the] language of the trial court, the conclusion is inevitable that, if this accident to the plaintiff was really by a. shock of electricity from thi,| controller, it came suddenly and from a hidden cause, which no cart or foresight could guard against, and for which the defendant is no| liable.
This reasoning determines other questions raised on appeal. I| the condition of the controller was consistent with the exercise of due care, it is immaterial that to have used solder, instead of screws might have secured a more permanent and a safer grounding, or thal paint, a nonconductor, was used on the inside of the brace where i| was attached to the casing of the controller, or that the court ex| *513eluded testimony as to the liability of the brass screw connecting the controller with the ground wire to become corroded, or that here the motoneer would be subject to additional danger if described insulating devices were omitted, or if other motormen using similar controllers received shocks, or the like.
Affirmed.
ELLIOTT, J., took no part.[November 29.]