Three questions are presented by this appeal: (1) That there was error in imposing tbe burden of proof on tbe defendant on tbe second issue. (2) That if tbe burden of proof was on tbe defendant, it was by reason of tbe presumption arising from proof that tbe defendant destroyed tbe property of tbe plaintiffs by fire, and that this was a presumption of law and not of fact; and that when evidence was offered rebutting tbe presumption, it was error to leave tbe question to tbe jury, in tbe absence of other evidence of negligence, and that it ought to bave been decided as matter of law by tbe court. (3) That it was error to refuse to nonsuit tbe plaintiffs on all tbe evidence.
(1) Tbe learned counsel for tbe defendant urges with much force on the consideration of tbe Court several cases in our own reports bolding that tbe burden of proof is on tbe plaintiff as to negligence, and that while tbe duty of proceeding with tbe evidence may shift from one party to tbe other, tbe burden of tbe issue does not shift; and be insists, on tbe authority of these cases, that there was error in bolding that tbe burden on tbe second issue was on tbe defendant.
An examination of these decisions will show that iri all of them one issue was submitted to tbe jury to determine tbe lia*423bility of tbe defendant, and that this issue embraced two facts: the origin of the fire, and the negligence of the defendant..
In the case before us these facts were to be settled by separate issues, and in this is to be found the distinction between the cases relied on and the one under consideration.
The first issue establishes {he fact that the defendant destroyed the property of the plaintiff by fire, and from this fact alone the presumption arises that the defendant was negligent. Ellis v. R. R., 24 N. C., 138; Lawton v. Giles, 90 N. C., 380; Manufacturing Co. v. R. R., 122 N. C., 881; Hosiery Mills v. R. R., 131 N. C., 238; Lumber Co. v. R. R., 143 N. C., 324; Deppe v. R. R., 152 N. C., 82; Kornegay v. R. R., 154 N. C., 392.
These authorities place the burden on the defendant to rebut the presumption of negligence arising from proof connecting it with the origin of the fire, by evidence which will satisfy the jury that the engine was properly equipped, that competent men were in charge of it, and that it was prudently operated; and, necessarily, the burden of the issue embracing these facts alone is on the defendant.
(2) The prayers for instruction tendered by the defendant require a consideration of the nature of the presumption in cases like this, because if this presumption is evidence in behalf of the plaintiff, the evidence of the defendant is not uncontra-dicted, as the instruction required the judge to charge.
It may be well to analyze the instructions before discussing them. They require the judge to decide that the evidence of the defendant is uneontradicted, and that, if believed by the jury, it is sufficient to establish the fact that the engine was properly equipped and was prudently operated by competent employees.
In many jurisdictions it is held that the presumption of negligence arising from proof that the defendant set out the fire is one of law; and generally, where this conclusion is reached, the courts approve the view contended for by the defendant, that it is the duty of the court to pass on the sufficiency of the rebutting evidence as matter of law.
*424This position is also supported by the ease of Williams v. R. R., 130 N. C., 116, in whieb it was held to be error to refuse to give an instruction substantially like those requested by -the defendant.
On the other hand, when the presumption is treated as one of fact, the rule usually obtains that the evidence must be submitted to the jury; who must pass on its sufficiency; and with the exception of Williams v. R. R., supra, our Court has held the presumption to be one of fact.
In Cox v. R. R., 149 N. C., 118, Justice Walker, speaking for the Court, says: “The presumption is one of fact and not law. Evidence that the sparks were emitted from the engine and that they set fire to the timber made a prima facie case for the plaintiff, but only to the extent of being evidence sufficient to carry the cáse to the jury and to warrant a verdict in favor of the plaintiff, if the jury should find the ultimate or crucial fact that the fire was caused by the defendant’s negligence.”
In Deppe v. R. R., 152 N. C., 82, Justice Manning, after stating the duty imposed on the defendant, says: “If the defendant can show at the trial that it ‘had used all those precautions for confining sparks or cinders’ which are approved and in general use, and the jury shall so find the fact, the trial judge will instruct them to answer the issue of negligence ‘No,’ provided the precautions were used by a competent and skilled engineer, in a careful way. Rule 1, in Williams v. R. R., 140 N. C., 623; Knott v. R. R., 142 N. C., 238.”
Note that after the rebutting evidence is introduced by the defendant, it for the jury to find the fact.
These cases and others to the same effect áre cited with approval in Kornegay v. R. R., 154 N. C., 392, where the principle is stated as follows: “When it is shown that the fire originated from sparks which came from the defendant’s engine, the plaintiff made out a prima facie case, entitling him to have the issue as to negligence submitted to the jury, and they were justified in finding negligence, unless they were satisfied, upon all the evidence in the case, that, in fact, there was no negligence, but that the defendant’s engine was equipped with a *425proper spark arrester and bad been operated in a careful or prudent manner. Williams v. R. R., 140 N. C., 623; Cox v. R. R., 149 N. C., 117.”
Tbe reasons for tlxo rule, and its justice, are nowbere better stated tban by Chief Justice Smith, in Aycock v. R. R., 89 N. C., 329: “A numerous array of cases are cited in tbe note 'in support of eacb side of tbe question as to tbe party upon wbom rests tbe burden of proof of tbe presence or absence of negligence, where only tbe injury is shown, in case of fire from emitted sparks. "While tbe author favors tbe class of cases which impose tbe burden upon tbe plaintiff, we prefer to abide by tbe rule so long understood and acted on in this State, not alone because of its intrinsic merit, but because it is so much easier for those who do tbe damage to show tbe "exculpating circumstances, if such exist, tban it is for tbe plaintiff "to produce proof of positive negligence. Tbe servants of tbe company must know and be able to explain tbe transaction, while tbe complaining party may not; and it is but just that be should be allowed to say to tbe company, ‘You have burned my property, and if you are not in default, show it and escape responsibility.’ ” Tbe note referred to is one to R. R. v. Schurty, 2 Am. and Eng. Enc. E. E. Cases, 271.
Tbe presumption is one of fact, and is itself evidence of negligence, and tbe evidence of tbe defendant in rebuttal of tbe presumption is as to facts upon which tbe decision of tbe issue depends, and there would seem to be no reason for excepting evidence of this character from tbe statute which forbids tbe judge from expressing an opinion on tbe facts or as to tbe weight of tbe evidence. If it should be held that tbe defendant was entitled to tbe instructions prayed for, tbe duty would be imposed on tbe judge to decide that there were no contradictions in tbe evidence of tbe defendant, that tbe witnesses were worthy of belief, and that tbe evidence was sufficient and satisfactory, which are matters committed by tbe law to tbe jury and not to tbe judge.
We conclude, therefore, that, assuming there were no contradictions in tbe evidence of tbe defendant, and that, if believed, *426it established the facts that the engine was properly equipped and was in charge of competent employees, and was prudently operated, this evidence cannot be said to be uncontradicted, and it was for the jury to pass on its weight. The evidence was contradicted by the presumption, which was some evidence of negligence.
We do not think Williams v. R. R., supra, is in line with the other decisions of this Court, and we must decline to follow it.
There are, however, other valid reasons for sustaining the ruling of his Honor.
There was some evidence of defects in the spark arrester, coming from the witnesses for the defendant. J. E. Bissett, master mechanic of the defendant, testified that there were patches on the wire netting, and that the covering of the manhole had long openings in it instead of square ones. He also said: “The covering of the manhole was in general use before they had adopted this wire netting, and we discarded that- — the master mechanics’ convention did — because the sparks would get hung in there and make a solid mass of it and the engine would get choked and the flame would come out the furnace door when it was opened and they would have to go in there and knock it, so they stopped this same netting to cover the manhole with. The reason those manholes with the long openings, instead of the square, were used, was that the S. A. L. had in stock a quantity of them, and we used them on the manholes to fill the bill, because there is enough opening, with what is in there, to give the engine draught enough to steam with.”
It was permissible to argue from this evidence that the spark arrester in use was old and dilapidated, and that it had been condemned by the convention of the master mechanics.
It is also doubtful if any evidence was introduced that the engine was properly operated.
Two engines of the defendant passed the place of the fire within a short time of each other, one being No. 746 and the other 752.
J. M. Stoker, engineer, was the only witness examined as to the operation of No. 746, and he says nothing as to how the *427engine was being managed, except tbat be was running about thirty miles an bour wben be passed tbe place of tbe fire; and tbe only evidence on tbis point as to engine No. 752 was tbat of N. E. Yaugban, engineer, wbo said be was sitting on tbe rigbt-band side of tbe engine, and was running at from thirty to thirty-five miles an hour. No inquiry was made of either as to what be or bis fireman was doing, or whether or not sparks escaped from tbe engine be was in charge of, as it passed tbe property of tbe plaintiff. In tbe absence of such evidence from witnesses wbo knew tbe facts, tbe jury might well infer tbat they were silent because a disclosure would be hurtful.
(3) If we are correct in our conclusion tbat tbe burden was on tbe defendant on tbe second issue, it-follows tbat there was no error in denying tbe motion to nonsuit, if there was evidence to support a finding in favor of tbe plaintiff on tbe first issue-as to tbe origin of tbe fire.
In our opinion, there was sufficient evidence to support tbe verdict.
Tbe lumber plant which was burned was situated near tbe right of way of tbe defendant. Engine No. 746 passed tbe lumber plant about 3:10 p. Mv and engine No. 752 about 3:30 p. m. Tbe fire occurred on Sunday, and several witnesses, wbo bad tbe opportunity to see, testified tbat they saw no smoke or other evidence of fire before tbe engines passed, and that tbe plant was burning within fifteen or twenty minutes after tbe passing of tbe last engine. Tbe engineer on engine No. 752 testified tbat be did not notice any smoke as be passed tbe plant, and one of tbe plaintiffs testified tbat be was at tbe plant about 1 o’clock p. M. on Sunday, and saw no evidence of fire, and tbat be was in tbe boiler-room tbe night before at 11 o’clock, and there were then only a few sparks in tbe back end of tbe boiler well.
Another witness for tbe plaintiffs, Mrs. Vick, testified tbat she noticed sparks, cinders, and heavy smoke coming from tbe train.
If tbis evidence is true, there was no fire about tbe premises before tbe engines passed; sparks escaped from tbe engine, and *428within fifteen minutes thereafter the property of the plaintiffs was on fire; and it was not unreasonable to conclude from these facts that the property of the plaintiffs was set on fire and burned by sparks from the defendant’s engine.
We have examined all of the exceptions appearing in the record, and find
No error.