• Action of trespass on the case to recover damages for the wrongful death of jfiaintiff’s intestate, alleged to have been caused by the negligence of defendant. After plaintiff had introduced all his evidence, the court sustained a motion to exclude it, and directed a verdict for the defendant; and plaintiff obtained this writ of error. The principal question is: Is plaintiff’s evidence sufficient to support a verdict in his favor,' if the jury had so found? If it is, the court should have allowed the case to go to the jury.
Plaintiff’s intestate, Frank J. Findley, deceased, was employed as fireman on defendant’s railroad, and had made two or three'runs before his death. On the 30th October, 1909, the boiler belonging to engine No. 16, which deceased was firing, exploded near a station called Yankee Dam, on Elk Eiver, as the engine was making a northbound trip, drawing a train of freight cars. Plaintiff’s intestate, the engineer and another fireman were instantly killed.
*270It is averred that defendant was negligent in that it did no't use due and proper diligence to keep its engine and boiler in a reasonably safe condition; that it “negligently and carelessly permitted and suffered the said boiler to be weak, unsafe and insufficient, the sheets of the said boiler to be and remain insufficient to withstand the stress and strain to which they were necessarily subjected and it negligently, carelessly and knowingly permitted the bolts and staybolts of the said boiler, to be,, and remain weak, unsafe and insufficient, broken off, rusted and corroded, so that the same were not sufficient to hold the said boiler together, and to resist the stress necessarily placed upon them in the operation of the said boiler.” The facts averred, if proven, 'would constitute negligence, because the master’s duty to his servant requires him to provide his servant with reasonably safe machinery with which to work, and to use reasonable diligence to maintain it in a safe condition. The degree of diligence necessary to preserve a locomotive boiler in a reasonably safe condition, is a mixed question of law and fact for jury determination, and must be determined from the experiences of men familiar with the construction and proper treatment of such powerful and dangerous machinery, when in use, men who have knowledge of the purposes and durability of its several parts.
The law imposes the burden of proving negligence on plaintiff; and the evidence by which he seeks to prove it is wholly circumstantial, consisting of testimony of witnesses concerning the appearance of the staybolts 'which held together the firebox and the main portion of the boiler. A number of witnesses: testified that they saw the broken staybolts immediately after the explosion, and that the broken ends of them were rusted' and corroded, thus indicating that they were broken sometime before the explosion. Such evidence is not only proper, but apparently it is the only available evidence to support plaintiff’s allegations. No living witness knows the amount of steam pressure on the boiler at the time of the explosion. But a witness who fired the engine about a month before the explosion, testified that the safety valve was set to a pressure of 18(1 pounds. It is also proven by a witness who saw the engine from across the river, when it exploded, that it was moving at the-*271rate of eight or ten miles an hour. While it is true that no-presumption of negligence arises from the mere fact of the explosion of the boiler,—Hanley v. Railroad Co., 59 W. Va. 419, and Veith v. Salt Co., 51 W. Va. 96—still negligence may be-established by proof of facts which show that the boiler was allowed to become unsafe before it exploded.
It was the duty of the "defendant to protect its employees against the dangers of an explosion, by having its boiler continuously inspected, and repaired when necessary, by competent machinists. If its inspector was negligent, and his negligence-was the proximate cause of the death of plaintiff’s intestate, then defendant is liable, the duty to maintain the machinery in a reasonably safe condition being a duty which defendant could not delegate to another so as to relieve it from liability. It is what the law denominates a non-assignable duty. Johnson v. Railway Co., 36 W. Va. 73.
Harry Bernard, who had had several years experience in making,and repairing boilers, testified that it was a good practice, and the general custom, to wash out locomotive boilers, and test the staybolts once a month. Staybolts have small holes,, about one-eighth inch in diameter, drilled into them, longitudinally, far enough to pass beyond the inner surface-of the sheets forming the fire box, and when one breaks, which appears to be not an unusual occurrence, the water and steam in the boiler is forced through the hole. In a large locomotive-boiler there are about eight hundred of these staybolts, placed in rows about four and a half inches apart, each way. The firebox is built into the rear end of the boiler, the sheets of metal forming it being held in place, and strengthened by the stay-bolts extending to and connecting with the sheets forming the barrel of the boiler. The space between is filled with water and steam.
Witness Bernard testified that, if one staybolt is broken, more or less strain is shifted to the ones next to it; that a bolt does not break off suddenly but -begins to crack and breaks gradually; that the breaking is'caused by the vibration, contraction, and expansion of the metals; that one bolt may be found broken off and another next to it only cracked; that if a broken bolt is: not removed the broken ends will become corroded; that the-. *272custom is, when replacing a broken bolt, to take out the one next to it also, and if it is found to be cracked, to continue to remove them successively, in that row of bolts, until a sound one is reached, and that when a sound one is found it is an indication that the remaining ones in that row sy:e sound.; that the bolts usually start to break in the corners of the fire box; that they are liable to rust, but that they usually break off before they are much affected by rust; that, as a rule, they are not allowed to remain long enough to be weakened very much by rust; that a broken bolt is not an infrequent occurrence, and does not indicate that the boiler is unsafe; that if a bolt breaks, while the engine is on the road, it is usual for the engineer to plug the test hole in the bolt by driving a wire nail into it to prevent the flow of water and steam; that staybolts are tested at the machine shops, by getting inside the fire box and tapping on the end of the bolt with a hammer, first having removed the fire and washed out the boiler and allowed it to cool; that a broken bolt is easily detected; that good practice requires such tests to be made every thirty days, if a boiler is constantly used, and that, if proper repairs are made, he did not think a sufficient number of staybolts would break off in that length of time to weaken a boiler to such an extent as to cause it to explode.
J. L. Peters, a machinist who used to work in defendant’s railroad shops, testified that he knew engine No. 16, whiclr was the engine that exploded; that it had been in use on the road since 1905 or 1906; that sometime in the winter of 1908 or .1909, while working on the night shift, he remembers to have corked some of the broken staybolts in the boiler by riveting the ends so as to close'up the test holes; that he knew of no other workman stopping the holes in that way; that he was not told to close the leak in that way, but volunteered to do it; that the last general overhauling that he remembers engine No. 16 to have received was in 1906; that the locomotive boilers, in use by defendant, were supposed to be washed out, and the bolts inspected every twelve days; that defendant keeps a man at its shops for the purpose of making the tests, and that he has seen him do it by getting in the fire box and tapping on the ends of the staybolts with a hammer.
H. O. Droddy testified that he fired engine No. 16 about a *273month or so before tbe explosion; that some of the staybolts leaked then "on the left side inside of the cab and some leaking on the outside., and some on the right side”; that wire nails were driven in the holes to stop the leaks. Droddy saw the broken parts of the boiler on the day of the explosion, and, in his testimony in relation to the appearance of the broken stay-bolts, says: "Some had the appearance of old breaks and some new.” * * * * "Several of them were rusty. Very corroded over the ends, while others were fresh broken.” That of the number that were corroded over the ends as near as he could tell there were "twelve or fifteen — something like that.” He also testified that, when he was firing engine No. 16, lie saw Mr. Rogers, the engineer, try to cork some of the leaking stay-bolts, on the road, at Clay Court House. But does not state' when that was.
J. W. Boggs also saw the broken boiler shortly after the explosion, and in his testimony concerning the broken bolts, which, had nails or metal in the test holes, said: “I never examined, them all. I counted somewhere in twenty. I would not say positive. I didn’t count near all of them.”
There is other evidence, similar in character. But all of' plaintiff’s evidence was excluded on the ground that it did not, prove negligence. This was error. Being unexplained and un-contradicted, the evidence was sufficient proof to warrant the jury in believing that the explosion was due to the large number of broken staybolts which it was the duty of defendant’s employees in the machine shops to have replaced with sound ones. If they were negligent in that respect, defendant is liable. It matters not how skillful they were, or how frequently the engine' was overhauled, if the explosion is due to the failure to make-proper repairs. Because the duty to provide reasonably safe machinery is an obligation from which the law does not relieve the master, it is one of his non-assignable duties.
In view of the testimony of Bernard, that a broken staybolt can be easily detected, the testimony of other witnesses, as to the large number of staybolts that were plugged with nails, and the broken ends of which were corroded, and appeared to be old, signifies negligence of the inspectors, and tends to prove such a defect in the boiler, existing before the explosion, as could *274have been discovered and remedied by tbe exercise of reasonable diligence.
On cross-examination of witness Droddy, defendant’s counsel produced two papers which, being admitted by Droddy to have been signed by him, were read to the jury for the purpose of impeaching his testimony. The papers were in the form of questions and answers thereto made by the witness shortly after the accident, relating to what he knew.concerning it, he having fired the engine a month or so before the explosion, and having seen the explosion from his father-in-law’s home on the opposite side of the river. The statements were not sworn to, nor does it appear who propounded to Droddy the questions. But it does appear that, at the time the paper was signed, Droddy was in the service of defendant as fireman; that John Emmert was assistant general manager of defendant’s road, and that Mr. Kal-baugh was superintendent of its motive power; that Mr. Emmert had written to the witness, a short time after the accident, to come to his office in G-assaway; that he did so; and on that occasion, the written statements referred to in the following question which the court refused to allow witness to answer, were signed by him: "State what, if anything, was said there by Mr. Emmert, at the time you had the talk with him, or by Mr. Kal-baugh, at the time you had the talk with him in the private car, in regard to a statement of what you might know about that explosion. What, if anything, was said in the way of advice, caution or direction about your statement?” Counsel for plaintiff stated that, if witness was allowed to answer, he would say, "that he was told that trouble was apt to grow out of this accident and that he had better be careful about the statement that he had made and was going to make, and that, as he was in the employ of the company and had a family to care for, he had better be careful as to his statements and conduct.” The court should have permitted witness to 'answer the question. The paper was offered to impeach his testimony, and witness had a right to explain his motive for ■signing it. The matter related to his credibility, a matter of which the jury were the judges, and before they could fairly pass on it, they were entitled to hear witness’ explanation for having signed a previous statement, so apparently inconsistent with his testimony.
It was ■ also - error, for the same reason, to refuse per*275mission to answer the following question, relating to the same matter: “Q. State whether yon made the answers.or any of them, which are written in typewriter upon this paper?”
Bill of exceptions No. 5. Witness W. H. Belknap, who saw the broken parts of the boiler after it had been brought to Gass-away, on the day after the explosion, was asked what was the appearance of the ends and sides of the broken staybolts,^ and replied : “Well, the appearance of them were old and rusty looking to me.” The court struck out this answer, on motion of defendant’s counsel. This was error. True, this witness was not an expert, but the appearance of broken iron that has been long exposed to the action of air or water, as compared to its appearance when fresh broken, is a matter of common knowledge. How it appeared to witness, is how it really was, so far as it concerns the value of his testimony. And if his answer involved the expression of an opinion by the witness, it was in relation to a thing which he saw and was trying to explain to the jury. The rule is, that a non-expert witness may be allowed to express his opinion in connection with the -facts on which it is founded, when the matter concerning which he has testified can not be reproduced, and made clear, to the minds of the jury. “In such case the witness testifies as to the present conviction of his own mind as to an actual fact, though deduced from circumstances which can not be made palpable to others.” 12 A. & E. E. L. (2nd ed.) 488. In State v. Welch, 36 W. Va. 690, a non-expert witness was allowed to state that in his opinion a stain, seen .by him, was a blood stain. The appearance which the broken bolts presented to witness was, necessarily, his opinion of their actual condition; and, in view of the common knowledge of all men in respect to such things, he could not better explain to the jury what he saw, than by telling how it appeared to him. It would appear to others as it did to him. This identical question arose in a similar case, decided by the supreme court of Illinois,- 210 Ill. 140, and it was there held that: “Non-expert witnesses, in an action for damages caused by a boiler explosion, may be allowed to testify whether or not breaks in the stay-bolts of the boiler had the appearance of old or new breaks, in connection with the facts, so far as they can be described in words, on which their conclusions are based.”
The court likewise erred, and for the same reason, in refusing *276tb allow answers to be made to similar questions asked of the same witness, set forth in plaintiffs bills of exceptions, Nos. 6 and 7. He was also asked if he was “able to determine if the bolts were recently broken, or if they, or some of them, had been broken before the explosion.” Answer to this was properly refused, because it called for witness’ conclusion or opinion in regard to the very issue to be tried by the jury.
The court properly excluded the following answer by witness Belknap, to a question in relation to the number, location and condition of the broken bolts that he saw, viz.: “Well, they looked in bad condition to me.” This answer implies that it was witness’ opinion that the bolts were in bad condition before the explosion. It could not refer to the condition of them, produced by the explosion, because that was not germane to the point that was then the subject of inquiry. For the same reason, the court properly struck out the following answer made by witness J. W. Boggs,- to a similar question relating to the appearance of the broken . bolts, viz.: “Well there were several bolts broken or rusted off, or burned oif, or something. I don’t know how they got off, but they had been off for some time.” Witness could describe the appearance of the broken ends of the bolts, whether rusted or fresh broken, but it was the province of the jury to determine how long they had been broken. It was also .proper to strike out the same witness’ answer, set out in plaintiff’s bill of exceptions No. 12, in which, speaking of the bolts, witness says they “were badly broken,” they “had been badly rusted and burned off.” The language implies that,-in the opinion of the witness, a large number of bolts had been broken off for a long time. He was not an expert, and it would seem from his statement, that some bolts were burned off, that he had but little knowledge of the construction of a lo- - comotive boiler. The space through which the bolts passed was filled with water and steam, and it was not possible for them to burn off. His testimony, that some of the bolts were plugged with nails,- was proper evidence. It sufficiently appears that the purpose in driving nails in the holes in the bolts, was to stop-a leak, and that a leak indicated a broken or cracked bolt. It is; therefore, a natural and fair inference that all the plugged bolts, were broken before the explosion.
It was proper not to permit Bernard, the expert witness, to *277answer the question, whether or not it would be “exercising reasonable care” to weld or plug the test holes in the staybolts. That was a matter for the jury to decide, upon proper instructions by the court. It was one of the very issues involved.
There are a number of other exceptions taken to the ruling of the court upon similar questions of evidence, but we think what we have already said amounts, practically, to a decision of all such questions raised, and will enable the court, on a retrial of the case to avoid the commission of error.
There is, however, another assignment, involving the question of after-discovered evidence, which deserves consideration Plaintiff moved for a new trial on the ground of after-discovered evidence, and in support thereof tendered his affidavit, in which he states that on the day of the trial, and after the -verdict was rendered, he “was informed” that two other witnesses,- naming them, were present shortly after the explosion, and examined the broken parts of the boiler. He then states what he is advised tliose witnesses will state. Affiant does not produce the affidavit of either of the witnesses, or of his informant; no cause is shown for failure to produce affidavits of the witnesses themselves. ■ Moreover, the newly discovered evidence is only cumulative. The affidavit was-clearly not sufficient. State v. Stowers, 66 W. Va. 198; State v. Gebhart, 70 W. Va. 232; and Jacobs v. Williams, 67 W. Va. 378.
■ We are of the opinion that the evidence was sufficient to entitle the jury to pass on the question of defendant’s negligence, and that it was error to direct a verdict to be found in its -favor. We, therefore, reverse the judgment, set aside the verdict and remand the cause for a new trial.
Reversed. New Trial Granted.