delivered the opinion of the court.
This is the sequel to the case of Cromer’s administratrix against the Norfolk & Western Railway Company, reported in 99 Va. 763, 40 S. E. 54. At the first trial there was a verdict and judgment for the plaintiff, which judgment, on writ of error to this court, was reversed, and the case remanded for a new trial.
At the second trial the plaintiff again prevailed, and recovered the judgment now under review.
The instructions given the jury conformed to the opinion of this court on the first appeal, and the errór n6w assigned is the refusal of the trial court to set aside the second verdict off the .ground that it was contrary to-the law and evidence.
*669In essentials, the evidence on both trials was the same.
It appears that on Monday, January 8, 1900, a west-bound passenger train, on which plaintiff’s intestate, Cromer, was fireman, arrived at Pulaski behind time, and, while running at a high rate of speed, collided with some freight cars, which had escaped from a siding, upon which they were stored, to the main track, and Cromer was killed.
The siding in question extends in a westerly direction from the place of accident, by the Pulaski Iron Purnace, to another point on the main line. It further appears that on Saturday, before the accident, twelve cars loaded with ore and coke for the furnace were stored on the siding, with brakes fastened and in safe condition.
As was said by this court on the first appeal: “The following occurrence is of interest as tending to show the sufficiency, of the brakes to control those cars. On Saturday preceding the accident, the employees of the company charged with that duty were putting cars in upon the siding some of which were loaded with coke, when they came in contact .with cars laden with ore standing towards the east end of the siding. The coke cars, which were being pushed, and the ore cars, which were at rest, did not couple, and the latter were put in motion by the jar. A brakeman sprang from the car upon which he was standing, overtook the ore cars, seven in number, which were moving off, applied brakes sufficient to stop them, and then at least two more brakes, out of abundant caution.
“It would seem that brakes which were sufficient to stop cars when in motion would be ample to hold them when at rest.”
Just how these cars were set in motion on the evening of the accident is wholly a matter of conjecture.
The plaintiff introduced the yard engineer, who stored the cars on the siding, and he propounds two theories on the subject—namely, that the brakes had been either tampered with, or that the cars were started by the impact of twenty-two or *670twenty-three loaded cars which were brought on the siding from the west by the employees of the company. In support of the latter theory, it appears that a few moments prior to the accident, after the yard engineer had left the siding, and while he was in the switch office, he heard these shifting cars come in contact with the stationary cars. The reasonable inference, therefore, would seem to be that the latter were set in motion as a result of that impact. And that inference is strengthened by the incident of Saturday, referred to in the opinion of the court.
So far as the liability of the company is concerned, however, it is immaterial which theory is adopted. If the brakes, which were shown by experience, as well as by direct evidence, to be amply sufficient to hold the cars in position, were tampered with, the company would, of course, not be responsible; and, if the cars were displaced and sent adrift by contact with the shifting cars, it was due to the negligence of fellow servants of the deceased, for whose negligence the company was not liable under the then existing daw. Norfolk & Western Ry. Co. v. Nuckols’ Adm’r, 91 Va. 193, 21 S. E. 342. Therefore, upon neither theory has actionable negligence been brought home to the company.
It likewise appears that six months prior to the accident there was a derailing switch on the siding near its eastern terminus, the presence of which, it is insisted, would have prevented .the accident, and the removal of which constitutes negligence on the part of the company.
The same contention was urged at the first trial and before this court on appeal. With respect to it the court said:
“In view of the evidence in the case tending to show that the cars on the siding were provided with all the appliances necessary to keep them stationary, and that these appliances and all the rest of the machinery were in good order, it was error to instruct, or to assume in an instruction, that the duty of or*671dinary care, which the defendant owed to its servant, could only he met by a derailing switch to prevent the moving of cars from the siding to the main track.
“It is the duty of the master to exercise reasonable care for the safety of his servant, but he is not bound to provide the latest inventions or the most newly-discovered appliances. He is not bound to use more than ordinary care, no matter how hazardous the business may be in which the servant is engaged.”
These observations are as applicable to the evidence upon the second trial as they were to that on the first trial, and are conclusive in regard to the present contention.
Courts and juries cannot dictate • to railway companies a choice between methods, all of which are shown to be reasonably adequate for the purposes intended to be subserved. Thus to subject them to the varying and uncertain opinions of juries in questions of policy, and to substitute the discretion of the latter for their discretion, would be wholly impracticable, and would prove alike disastrous to the companies and the public. Tuttle v. Ry. Co., 122 U. S. 189, 7 Sup. Ct. 1166, 30 L. Ed. 1114.
In order to hold a master liable for injuries sustained by a servant while engaged in his employment, the burden rests upon the servant to prove affirmatively the negligence of the master, or a state of facts which warrants an inference of negligence, and that such negligence was the proximate cause of the injury. To justify a recovery in such a case, the evidence must show “more than a probability of a negligent act.” The existence of negligence must not be left entirely to conjecture, and courts cannot uphold the tentative conclusions of juries, based upon no sure grounds of inference. C. & O. Ry. Co. v. Sparrow’s Adm’r, 98 Va. 640, 37 S. E. 302.
Again, it is to the proximate cause, and not to the remote cause, that courts must look in determining the rights and liabilities in this class of cases.
*672It was urged in argument that this is a case of concurring negligence. But that doctrine cannot he invoked under the evidence in this record. It was held on the former appeal that concurring negligence of a master and servant, in order to render the master liable, must be simultaneous, operative, and effectual at the time of-the accident, and must not stand in the relation of remote and proximate cause to the event, for, if they so stand, and the servant can, by the exercise of ordinary care, avoid the effect of the master’s negligence, there is no liability on the master.
It is.believed, however, that a brief consideration of the circumstances immediately attending the accident will conclusively show that it was proximately due to the negligence of the engineman in disregarding a plain duty imposed upon him by the rules of the company, negligence in which Oromer participated, to the extent, at least, of not keeping a proper lookout on reaching the yard at Pulaski.
As remarked, the accident occurred within the yard limits, near the eastern junction of the siding with the main track. The profile map introduced in evidence shows a slight descending grade from the summit of the siding, in an easterly direction, to the bottom on the main line near the junction, at which point the ascending grade would stop a loose car within a short distance in the yard, and within the jurisdiction of the yard force.
It likewise appears that a car escaping to the main track would pass through a switch provided with automatic switch signals, which, when open display a red light, and when closed a white light (the former signifying danger; the latter, safety); and when, from any cause, these signals are obscured, their absence is a notification of danger.
The eastern limit of the yard is about one-half mile from the scene of the accident, and the intervening track is perfectly straight, and the view unobstructed.
*673Ordinarily, an accident from collision, under the physical conditions adverted to, would he most unlikely to occur, and the conclusive inference is that the casualty in question would not have happened if the engineman and fireman had observed the rules promulgated by the company for their guidance and protection.
These rules require that, “when within the limits of the various yards, all trains must be run with great care, and under the control of the engineman.”
“Switching engines will have the right to work within yard limits, upon the time of second and succeeding class trains, and also upon the time of delayed first-class trains, but must clear the track immediately upon their arrival. The main track must be kept clear for first-class trains that are on time. ‘First-class’ trains means passenger trains, arid ‘second-class’ trains means freight trains.”
Bule 164, which is especially applicable to firemen, reads as follows: “When running upon the road they must keep a constant lookout ahead when not engaged in firing, and give notice to the engineman of any signals or indications of danger. If the engineman has to look away from the track in front for any reason, the fireman must maintain the watch until the engine-man can resume it. They will not put coal in engines when coming into stations or at such other points as safety requires that they keep a lookout ahead.”
That these salutary rules were set at naught by the engine-man on the evening of the accident cannot be denied. He approached the scene of disaster, under existing conditions, at a reckless rate of speed. His train was an hour and a half late, and he, therefore, knew by the rules that he was not entitled to the right of track, and was likely to encounter obstructions within the precinct of the yard. He also knew that it was his duty to keep a constant lookout ahead, and to run his train with great caution, and keep it under control. So far from running *674with the care required, and reducing the speed so as to bring the engine under control, the undisputed evidence is that he maintained a speed of at least thirty miles an hour until his train was actually in collision with the drifting cars. Had the rules been obeyed in particulars referred to, it is inconceivable that the engineman and fireman would both have failed to discover either the freight cars on the track, or the switch signals, or the absence of such signals, all or any of which would have warned them of danger in time to stop a train under control, and thus to have averted the accident.
Erom the foregoing statement, it plainly appears that the direct and proximate cause of the accident was the negligence of the engineman, in which the fireman participated, and for their negligence the company is not responsible.
Eor these reasons the judgment complained of must be reversed, the verdict of the jury set aside, and the cause remanded for a new trial to be had, not in conflict with the views expressed in this opinion.
Reversed.