This action was brought to recover the value of a warehouse and its contents, which on May 2, 1895, were destroyed by fire caused by sparks emitted by a locomotive of the defendant. The plaintiff’s building stood about twenty feet from the defendant’s track. Within a very few minutes after the passage of the train a fire was observed on the roof of the warehouse, and at the same time another fire was discovered burning on the bank adjacent to the track of the defendant, further south. There had been no fire in or about the warehouse on that day and when the fire was first observed it had not burned through the roof to the interior of the building. Evidence was given showing that the defendant’s engines frequently emitted large quantities of sparks setting fire to grass and fences at a distance of sixty or seventy feet. It was proved that on the twenty-third of March, previous to the fire, the same locomotive which caused the fire in the *350plaintiff’s building had thrown out sparks or cinders, some of them as large as pea beans, which started fires at a distance of four or five rods from the railroad track. An expert connected with the Baldwin Locomotive Works testified as to the appliances used as spark arresters, their operation and' effect. He stated that with the appliances now in use the sparks emitted are not of sufficient size nor sufficiently heated to ignite anything along the road. Two locomotive engineers were examined on the same subject. The following questions - were put to them, to which the defendant’s objections were sustained and the plaintiff excepted : “ Q. What do you say as to the possibility of a spark or cinder the size of a pea bean being thrown through the netting or spark arrester such as you have described ? Q. What do you say as to the possibility of a spark or cinder being thrown out through a spark arrester such as you have described, in good condition, capable of setting a fire at a distance of fifty or seventy-five feet from the railroad track ? ” At the conclusion of the plaintiff’s case the complaint was dismissed.
It is conceded, both by the learned Appellate Division and by the counsel for the respondent, that the evidence was sufficient to justify the jury in finding that the fire in plaintiff’s warehouse was caused by sparks from the defendant’s locomotive. This, however, was not sufficient to authorize a recovery against the defendant, and the latter “ could not be made liable for the destruction of the house upon the adjoining lot, except upon proof of negligence in the management or condition of its engines. The action in such a case is based upon negligence, and a railroad company cannot be made liable for the unavoidable or usual consequences of the proper operation of its road to adjacent property.” (Flinn v. N. Y. C.& H. R. R. R. Co., 142 N. Y. 11.) But while it was necessary for the plaintiff to affirmatively establish negligence on the part of the defendant, either in the condition or in the operation of its engine, for which the mere occurrence of the fire was not sufficient, it was not necessary that he should prove either the specific defect in the engine or the particular act of *351misconduct in its management or operation constituting the negligence causing the injury complained of. It was sufficient if the plaintiff proved facts' and circumstances from which the jury might fairly infer that the engine was either defective in its condition or negligently operated. The emission of sparks unusual in quantity or character, or of an extraordinary size, such as would not be emitted from well-constructed locomotives in proper repair, would justify the jury in inferring negligence, and though not shifting the burden of proof, would cast upon the defendant the duty of explanation. (Field v. N. Y. C. R. R., 32 N. Y. 339; Bedell v. Long Island Railroad Co., 44 N. Y. 369 ; Flinn v. N. Y. C. & H. R. R. R. Co., supra; McCaig v. Erie Railway Co., 8 Hun, 599.) The plaintiff’s proof showed that defendant’s engine had thrown out sparks the size of a pea bean .a distance of several rods sufficiently heated after being carried that distance to set grass and fences on fire. The excluded questions put to the witnesses, had they been answered, might have shown that sparks or coals of the character described would not have been emitted by locomotives having proper appliances and in good repair. Indeed, we do not see how the plaintiff could establish the defendant’s liability by circumstantial evidence in any other manner than that which he attempted to pursue; that is to say, first, by proof of the quantity, size and character of the sparks actually thrown out by the engine; second, by proof by experts that sparks of such character would not be emitted from an engine in proper repair. The witnesses to whom the questions were put were locomotive engineers and competent to testify to facts within their observation and experience. The propriety of similar questions was upheld in Brush v. Long Island R. R. Co. (10 App. Div. 535; affirmed, 158 N. Y. 742), and in Jamieson v. N. Y. & Rochaway Beach Ry. Co. (11 App. Div. 50; affirmed on opinion below, 162 N. Y. 630). We have to make this qualification as to the second of the two questions referred to. That question is objectionable, in that it calls for an opinion as to whether the heat in a spark thrown a *352distance of fifty to seventy-five feet from a locomotive in proper condition would ignite any tiling along the road. The liability of ignition would depend on the material or substance along the line of the road upon which the .sparks might fall as well as the conditions of those substances at the particular time. This was not a proper subject for expert evidence, at least from the particular experts on the stand as witnesses. Those witnesses, however, might properly testify as to’ the condition of sparks or cinders thrown to such a distance; whether, in their observation and experience, live or burning sparks would have been carried so far if the engine was in proper order. It is doubtful whether the defendant’s objection fairly raised the point in which we regard the question as objectionable. It is, however, not necessary to pass upon this, as we think the question whether a locomotive, in proper repair and condition, would have emitted sparks as large as those shown to have been thrown out by the locomotive which caused the fire, was clearly proper and its exclusion error. We cannot say that this error was harmless; on the contrary, we are of opinion that the evidence for the plaintiff, meagre as it was, if supplemented by the excluded testimony, would, in the absence of proof or explanation on the defendant’s part, require the case to be submitted to the jury.
The judgment of the trial court and the Appellate Division should be reversed and a new trial ordered, costs to abide the event.
Parker, Ch. J., Bartlett, Martin and Vann, JJ., concur; Gray and Wernerb, JJ., not sitting.
Judgment reversed, etc.