PECKHAM, J.

Among other questions of fact submitted to the jury, which if found in favor of the plaintiff a verdict for him might be based, was the question whether the system of spark arresters used by defendant was suitable and whether defendant was guilty of any negligence in using that system.

The jury were permitted by the charge of the learned judge to say that the system of spark arresters was not a proper one, and that it was not one of the most approved devices known for the prevention of the escape of sparks or cinders from the engine. The evidence was entirely uncontradicted and tended solely to one result, viz., that the straight stack spark arrester was at least as good as any system that was known. Some of the witnesses said that it was the improved spark arrester, as good as any spark arresters that were known to railroad men. It began to come into use about 1880 and was much used for freight trains. There was the straight stack without the extension. There were also some with an extension and there were diamond stack engines, but the latter were, perhaps, not quite so good as others. The straight stack without the extension was the kind of spark arrester that was used on the engine in question. The fact was testified to by witnesses for the defendant who were among its employees, but they were men who were not responsible for the system of spark arresters that was adopted or in general use upon defendant's road, and they were testifying upon a subject well known among all men familiar with railroad engines and the system of spark arresters in existence or adopted by different railroads, and if there had been in fact the least doubt upon the question regarding which they testified, evidence contradicting or explaining their testimony could have been easily adduced.

In the plaintiff's complaint there was an allegation that the defendant had not equipped or provided the engine with a kind of spark arrester that was reasonably safe and such as was in common use by the railroads of the state and had been for a long time to the knowledge of defendant, and which it was its duty to put upon and provide its locomotives with for the protection of property along the line of its road.

Upon the trial the plaintiff gave no evidence to sustain this particular allegation, and the only evidence in the case came, so far as this issue is concerned, from the defendant, and that was to the effect already stated. The real issue made by the plaintiff, and to which his evidence was directed, was that of negligence in regard to the condition of this particular spark arrester and its being out of repair and its meshes too large, so that sparks of an unnecessarily large size escaped therefrom and in fact set the fire in question.

The engine had just been equipped with this new spark arrester, the change having been made from the diamond stack kind. In the case of Flinn v. Central Hudson Railroad ( 142 N.Y. 11) the diamond stack spark arrester was used and the claim was made by the plaintiff in that case that the defendant should be held liable because it had not adopted the straight stack spark arrester, such as was used upon this engine. Here the claim seems to be just the opposite, and the jury was permitted upon the evidence in the case to find that the straight stack system was not a good one and the diamond stack should have been used. Thus, whichever kind may be used by a railroad company, it is open to the plaintiff to claim that it was the other that should have been used, and the question must, therefore, be submitted to the jury.

We think the court can take judicial notice of the fact that diamond stack and straight stack spark arresters are in very general use upon the railroads of the country and that they are both well-known systems for arresting sparks, while no system that has yet been invented can wholly prevent the emission of live sparks from an engine under certain circumstances. In the Flinn case we held that the defendant could not be found guilty of negligence in failing to introduce upon its road the straight stack extension (the kind actually used in the case at bar) within the time then existing since its invention, even assuming that the straight stack was an improvement on the old diamond stack form. Here the jury is permitted to find, in opposition to the whole evidence in the case, that the straight stack extension is not a proper system of spark arrester. We hold it was error to submit to the jury the question of defendant's negligence in the adoption of a proper system or kind of spark arrester.

The case is not within the principle decided in Cornish v. Insurance Co. ( 74 N.Y. 295). In that case it was held that the testimony of experts, although uncontradicted, is not conclusive, except in cases where none but experts are capable of determining the question. That was a case depending upon special circumstances, and the jury was called upon to judge whether the general principles governing somewhat similar cases, as testified to by the experts, were applicable to the case before them. It was a question as to what really increased the risk in an insurance against fire, and although a question of fact, insurance men who were engaged in the business of insuring property or procuring policies of insurance thereon, were called as witnesses to state whether the risk was in fact increased by permitting a building under the circumstances to remain unoccupied. Although the experts testified that the risk was thus increased, yet it was, of course, but a matter of judgment or opinion, and upon a question that the jury must itself decide. So, although the expert evidence was admissible upon that issue and was also uncontradicted, it was held by this court not to be conclusive.

The evidence in this case is of a much more comprehensive nature. The question was whether there was any other system better or as well known, and which, indisputably, was more suitable than the one used by defendant for the arresting of sparks. The witnesses testified that this particular system was in general use on many of the large railroads, and on some was almost exclusively employed, and that it, in fact, arrested sparks as well as any kind that was known. The plaintiff gave no evidence to the contrary, and these facts stood proved by the evidence of seemingly respectable, intelligent and credible witnesses, and the proof was uncontradicted.

We think there is so much difference in the kind of evidence given in the two cases as to render that of Cornish ( supra) inapplicable to the case at bar.

The error necessitates the granting of a new trial for the reason that we cannot possibly say that the jury based their verdict upon some other fact, and not upon a finding that the system of spark arresters was improper and inappropriate. As there must be a new trial, the question of the liability of the defendant for the burning of the hotel will again arise. We think the charge of the learned judge upon this part of the case was as favorable to the defendant as it could properly ask. The question was left as one of fact, under all the circumstances, as to whether the burning of the hotel were not the natural and direct result of the sparks from the engine. In this case the court committed no error to the prejudice of the defendant. The Ryan Case ( 35 N.Y. 210) should not be extended beyond the precise facts which appear therein. Even if correctly applied in that case, the principle ought not to be applied to other facts. (See Webb v. Railroad Co., 49 N.Y. 420; Pollett v. Long, 56 id. 200; Lowery v. Manhattan Railway Co., 99 id. 158; O'Neill v. Railroad Co., 115 id. 579.)

For the error above stated the judgment must be reversed and a new trial granted, with costs to abide the event.

All concur, except ANDREWS, Ch. J., not sitting.

Judgment reversed.