Opinion,
Mr. Justice Clark :This action was brought to recover damages for the destruction by fire of the plaintiff’s sash and door mill at Montgomery, in Lycoming county. The mill was situate between the Pennsylvania and the Philadelphia & Reading railroads; the former passing in front, and the latter in the rear of the mill. The plaintiffs allege that the fire, which occurred on the tenth day of August, 1888, was communicated from sparks emitted by the defendant’s engines. The fire was discovered about *4756 or 6.15 o’clock p. M., in the upper part of the ventilator, on the side next the defendant’s road. The ventilator was about thirty feet high, and was within twenty-two feet of defendant’s road.
The watchman testifies that he came on duty that evening about fifteen minutes before shutting-down time, and that the mill shut down at about 5.30 p. m. mill time, or 5.15 railroad time; that after he came on duty, and before the fire, two trains passed; the first a coal train, going north, drawn by an engine which he could not identify; and, about fifteen minutes later, a freight train, drawn by engine No. 72. The defendant’s evidence, however, showed that two other engines, drawing passenger trains, passed this point, one at 5.21 and the other at 5.22 p. m., neither of which engines was identified; indeed, it would seem that the plaintiffs did not know they had passed the mill until the fact was developed in the defendant’s testimony. The watchman testifies, further, that it was his duty to take notice of the engines as they passed, to see whether they threw fire from the stacks; that he did watch the engine in front of the coal train, and also engine No. 72, and that he saw no sparks; but that, as it was only six o’clock, and the sun was shining brightly, there may have been sparks emitted which he did not see. The only engine known, and identified was No. 72.
The defendant’s contention was that the fire occurred in the pit containing the shavings and débris of the mill, which was immediately underneath the ventilator, and from which the shavings, etc., were supplied as fuel to the furnace. There is a large volume of testimony bearing upon the origin and cause of the fire, upon consideration of which the jury found the fire to have been caused by sparks from the defendant’s locomotive engines.
The Philadelphia & Reading Railroad Co., at the time of the injury complained of, was an incorporated company, entitled to the right of way for its engines, etc., upon their track, as located in the rear of the plaintiffs’ mill. The company, in the proper use of its road, was therefore in -the lawful pursuit of a legitimate business, and if injury resulted to the plaintiffs, it is damnum absque injuria; the company cannot be mulcted in damages except upon proof of negligence: Frankford etc. *476Turnp. Co. v. Railroad Co., 54 Pa. 345 ; Phila. etc. R. Co. v. Hendrickson, 80 Pa. 182. No person is answerable in damages for the reasonable exercise of a right, when the act is done with a cautious regard for the rights of others, and where there is no ground for the charge of negligence, unskilfulness, or malice. For the ordinary risks, the landowner is compensated in the damages for right of way; negligence, therefore, is the gist of the action, and the burden of proof is upon the plaintiffs to establish it. And as all engines, whether provided with spark-arresters or not, emit sparks, the mere existence of a fire along the line of the road, caused by sparks from the company’s engines, is not enough to fasten upon the company the charge either of negligence or want of skill: Phila. etc. R. Co. v. Yeiser, 8 Pa. 366. In Jennings v. Railroad Co., 93 Pa. 340, this court in a per Curiam opinion, said: “ To hold that the fact of the fire having taken place was prima-facie evidence that the spark-arrester was defective, and therefore that the case ought to have been submitted to the jury, would be practically to hold railroad companies liable for all fires; for it is notorious that no spark-arrester has yet been invented to prevent all sparks; and a little spark may kindle as large a conflagration as a large one, it depending very much on the dryness or humidity of the atmosphere whether a spark will go out before reaching the ground, and whether what it reaches is in a condition to be easily ignited.” So, also, Phila. etc. R. Co. v. Schultz, 98 Pa. 344; Reading etc. R. Co. v. Latshaw, 93 Pa. 449.
Whilst any ordinary fuel may be used in a locomotive engine for the generation of steam, the exercise of this right is subject to the restriction that the latest improvements in its management in general use shall be applied to it: Frankford etc. Turnp. Co. v. Railroad Co., 54 Pa. 345. It is the duty of the railroad company, in the use of an engine, to use such reasonable precaution as may prevent damage to the property of others; hence, in Lackawanna etc. R. Co. v. Doak, 52 Pa. 379, where, although there was no direct evidence that the building was fired by the engine, or that sparks were emitted from it at the time, yet the building was near the railroad and was discovered to be on fire when the train passed, and it was shown that the engine had no spark-arrester, it was held that the *477question of negligence was properly submitted to the jury. The effect of this ruling was to establish the principle in Pennsylvania that in case of loss by fire, fairly attributable to sparks from a railroad company’s locomotive engine, the absence of a spark-arrester is prima-facie evidence of negligence on the part : of the company. It is the duty of railroad companies to adopt the best precautions against danger in general use and which experience has shown to be superior and effectual, and to avail themselves of every such known safeguard or generally approved invention to lessen the danger. But mechanical invention and skill have all provided a merely partial protection against the emission of sparks. The mere fact that sparks are thrown from the stack of an engine is not, therefore, evidence in itself of negligence. Where, however, sparks of large size are emitted, which, carried to a long distance, set fire to fields, fences, or buildings, it may, in the present condition of this branch of mechanical invention, well be inferred that the engine is not provided with a sufficient spark-arrester: Phila. etc. B. Co. v. Hendrickson, supra; Pennsylvania Co. v. Watson, 81* Pa. 293 ; Penna. etc. R. Co. v. Lacey, 89 Pa. 458 ; Phila. etc. R. Co. v. Schultz, 93 Pa. 341. Therefore, in an action for the recovery of damages for the destruction of a dwelling seventy-seven feet distant from the railroad, where it was shown that sparks were seen flying from engines to a distance of more than fifty yards, and fences and fields were set on fire in several places, about the same time and at considerable distance from the road, the question of negligence, it was held, should have been submitted to the jury. Although the company gave evidence to the effect that their engines were in good order, and were all provided with good spark-arresters, the unusual distance to which the sparks were borne, and the numerous fires they created, were held to be such evidence to the contrary effect as to have carried the case to the jury: Huyett v. Railroad Co., 23 Pa. 373.
Where the injury complained of is shown to have been caused, or, in the nature of the case, could only have been caused by sparks from an engine which is known and identified, the evidence should be confined to the condition of that engine, its management, and its practical operation. Evidence tending to prove defects in other engines of the company is irrelevant, *478and should be excluded: Erie Ry. Co. v. Decker, 78 Pa. 293. In the case cited, the house of the plaintiff, which stood near the track of the defendant’s railroad, was destroyed by fire on the sixth of March, 1872. The plaintiff alleged that the fire originated from sparks thrown from locomotive engine No. 458, belonging to the defendants, which passed his house about the time the fire commenced, and that the throwing of the sparks was from the negligence of the defendants in not having their apparatus in proper order. Mr. Justice Gobdon, in the opinion of the court, says: “ It appears from the evidence, and it was conceded in the argument, that the only locomotive that could have fired the premises in question was that numbered 458, in charge of Alfred Carpenter as engineer. It follows, therefore, that the condition of this engine and its management were all that were legitimately before the court. If it was properly constructed as to its furnace and smoke-stack, and was furnished with a spark-arresting grate of the proper character, the company would not be liable, though the building were burned by fire accidentally issuing from it: Lackawanna etc. R. Co. v. Doak, 52 Pa. 379. If, then, this engine was in a proper condition, it mattered not that every other engine owned by the company was without the proper appliances for preventing the ejection of coals and sparks. On the other hand, if this engine was dangerous in this respect, it was of no consequence that all others upon the road were safe. Such being the case, it is manifest that all evidence going to prove defects in engines belonging to this company, other than the one alleged to have produced the injury complained of, was irrelevant to the issue pending, and should have been excluded.”
So, in Albert v. Railway Co., 98 Pa. 316, where it appeared that the plaintiff’s loss, if indeed it was caused at all by the defendant’s negligence, was attributable entirely to the escape of sparks at a particular time from one of two particular engines, both of which were identified, evidence was held inadmissible on the part of the plaintiff, in order to prove defendant’s negligence, to the effect that sparks of unusual size had been emitted for some time prior to the fire by defendant’s engines generally. “ The evidence below,” said our Brother Paxson in that case, “ established the fact that, if the plaintiff’s property was de*479stroyed by fire communicated by defendant’s locomotive, it was done by engine No. 21 or engine No 126, and by no others. Hence, it is entirely clear that evidence that other engines, upon some other day, threw out an unusual amount of large sparks and live coals, was immaterial, and if received could only have confused and might have misled the jury; nor would it have been evidence to show that the spark-arresters on engines 21 and 126 were out of order.” That is to say, for the last sentence is perhaps a little obscure, the fact that other engines, at other times, threw out an unusual amount of large sparks and live coals, would not have been evidence to show that the sparkarresters on engines 21 and 126 were out of order. To the same effect is Jennings v. Railroad Co., supra; Annapolis etc. R. Co. v. Gavitt, 39 Md. 124; and other cases that might be cited.
Of course, the inquiry in all such cases is as to the existence or condition of the spark-arrester at the precise time of the injury ; but, in order to make this practicable by proof that it was defective, or threw out sparks of unusual size, a reasonable latitude must be allowed to show its management and operation both before and after. The evidence, however, must be confined to its operation at or about the time of the occurrence. In Phila. etc. R. Co. v. Schultz, supra, it was shown that every day for two weeks a particular engine had been observed to throw out quantities of unusually large sparks, and had fired property along the line of the railroad. In Albert v. Railway Co., supra, it was shown that both engines then in question had done this for some time before the occurrence. To the same effect, also, is Lehigh V. R. Co. v. McKeen, 90 Pa. 122. Testimony tending to show that other fires were set by the same engine about the same time, however, is the proper rule, and is undoubtedly competent: Boyce v. Railroad Co., 43 N. H. 627; Grand Trunk R. Co. v. Richardson, 91 U. S. 454.
But, when the loss or injury is shown to have been caused, or, according to the proof, may have been caused by sparks from an engine unknown and unidentified, or by one of several engines, some of which are unknown and unidentified, then the rule of evidence is necessarily somewhat enlarged. The burden of proof in all such cases, in the first instance, is upon the plaintiff to show that the fire in question was communicated from the defendant’s engines. “ It devolves upon the plaintiff *480to prove by a preponderance of tbe evidence that the fire was communicated by sparks or cinders from the railway engines. It need not be shown that any particular engine was at fault, but it will be sufficient if the fire is proved to have been set by any engine passing over defendant’s railway, and the evidence may be wholly circumstantial; as, first, that it was possible for fire to reach the plaintiff’s property from the defendant’s engines ; and, second, facts tending to show that it probably originated from that cause, and from no other: ” 8 Am. & Eng. Enc. of Law, 7.
And, although the rule is otherwise in England and in many of the states, in Pennsylvania, as we have said, the additional burden is upon the plaintiff to prove negligence in the construction or management of the engine. It is not required that the fact be established by direct or positive proof; like any other fact, it may be established by circumstantial evidence ; and, on account of the great difficulty in proving negligence in such cases, any proper evidence from which negligence may be inferred is sufficient to throw the burden on the defendant. “A slight presumption of negligence, however, raised by the plaintiff’s case,” says Mr. Wharton in his Law of Evidence, § 871, “is sufficient to throw the burden of disproving negligence on the defendant. It is a mistake, as has been elsewhere shown, to suppose that negligence can be only proved by positive and affirmatory evidence. There may be no direct proofs of negligence, yet the way in which an injury is done may be such that negligence is the most probable hypothesis by which it can be explained; and when this is so, the defendant must disprove negligence by showing that he exercised care.” In Thompson on Negligence, 159, it is said: “ The business of running railroad trains suggests a unity of management, and a general similarity in the construction of the engines. For this reason, and on account of the difficulty of proving negligence in these cases, as before pointed out, the admission of evidence as to other and distinct fires from the one alleged to have caused the injury is permitted. The rule is adopted in England, and prevails in all the states, with one, or possibly two, exceptions. More particularly, it may be stated as follows: That, in actions for damages caused by the negligent escape of fire from locomotive engines, it is compe*481tent for the plaintiff to show that, about the time when the fire in question happened, the trains which the company were running past the location of the fire were so managed, in respect to their furnaces, as to be likely to set on fire objects in the position of the property burned, or to show the emission of sparks or ignited matter from other engines of the defendant passing thé spot upon other occasions, either before or after the damage occurred, without showing that they were under the charge of the same driver, or were of the same construction as the one occasioning the damage.” The rule is more precisely stated in Shear. & Redf. on Negligence, § 675, as follows: “When the particular engine which caused the fire cannot be fully identified, evidence that sparks and burning coals were frequently dropped by engines passing on the same road upon previous occasions, is relevant and competent to show habitual negligence, and to make it probable that the plaintiff’s injury proceeded from the same quarter. If the engine which emitted the fire is identified, then evidence on either side as to the condition of other engines, and of their causing fires, has been held irrelevant, but not so if it is not fully identified.”
In our own case of Penna. R. Co. v. Stranahan, 79 Pa. 405, the evidence was that, between 2 and 8 o’clock in the afternoon, the plaintiff’s barn, which was about one hundred and fifty feet from the railroad, was diseoved to be on fire. Two trains had passed about noon. The fire appeared to have commenced at the fence on the road, and burned over the field to the barn. The sparks falling set fire in many other places along the road. The engine from which the sparks were alleged to have been thrown was unknown and unidentified, and the plaintiff proposed to show by a witness who lived nineteen miles distant on the line of the railroad, the extent to which the locomotives on that road going east, on or about the time of the occurrence, threw sparks from the smokestacks. The testimony was admitted. The witness testified that it was “ a common occurrence for the engines to throw sparks, and set fire for rods from the railroad track; they were from a pea to a walnut in size; it appeared worse sometimes than others; they were usually freight trains; sometimes passenger trains,” etc. The admission of this testimony was assigned for error here. In a per Curiam opinion, this court said: “ This was *482not a case where a certain engine had thrown out the sparks which set fire to plaintiff’s barn; but it was where the engine was unknown, yet the cause of the fire was clearly traced to the railroad track, and left the belief that some one of the engines of the defendant had emitted the coals which set the barn on fire. It therefore became necessary to establish the fact by such proof as rendered the belief a certain fact. This could be done, not by the proof that a certain engine emitted the sparks incessantly; for non constat that this particular engine had passed the plaintiff’s premises that day. Hence it was necessary to permit the party to show that the emitting of coals and sparks in unusual quantities was frequent, and permitted to be done by a number of engines.” In Gowen v. Glaser, 3 Cent. R. 109, the action was for damages for the destruction by fire of the plaintiff’s rags, which were scattered in a field adjoining the defendant’s road. The allegation was that they were set on fire by sparks from the defendant’s engines, but it was not known by what engine. The offer made was as follows: To show that several engines on this road had insufficient spark-catchers ; that the engines of this road had repeatedly set fire to property, and to vegetation along that part of the track, very shortly before and very shortly after this occurrence ; that sparks as large as a hickory-nut escaped in large quantities from the engines, causing these fires; that, after this fire, what remained of the rags, and what was saved, were spread on the field and watched day and night, and that they were set on fire repeatedly by the engines passing on this road. This offer was received to show by circumstantial evidence that the damage was done by some engine with an insufficient spark-arrester. The jury were to infer from the fact that many of the company’s engines, about the time of this occurrence, shortly before and shortly after, emitted sparks of unusual size and quantity, that they were without sufficient spark-arresters; and that, upon consideration of all the evidence, the injury complained of resulted from some one of the engines thus imperfectly constructed. The offer was subsequently enlarged by adding to it a proposition to prove, not that the whole number of defendant’s engines were defective, but that the defendant habitually used engines with defective spark-arresters. The offer, as a whole, was admitted, and in *483this court was assigned for error. In a per Curiam opinion, this court held that there was no error in the admission of this offer. In Railroad Co. v. Page, 21 W. N. 52, the action was for burning the plaintiff’s barn, one hundred fifty feet distant from the track. The evidence was that the company’s trains had passed the barn shortly before the fire broke out, emitting cinders, smoke, and small sparks about the size of a pea. There was no evidence, direct or circumstantial, to justify the jury in finding that the sparks were of any larger size. It was further shown that the wind was blowing from the track towards the barn, and that sparks had been known to have been blown that distance. It was not shown that any spark-arrester in use would effectually prevent the emission of sparks of this size. Whilst the evidence was, perhaps, sufficient to satisfy the jury that sparks from the engine had caused the fire, there was no proof of any defect in the spark-arresters ; on the contrary, it was shown they were in perfect condition. There was therefore no proof of negligence or mismanagement; and it was upon this ground that we said it would have been the duty of the court below, if a proper request had been made, to instruct the jury to find a verdict for the defendant. ■
The same rule of evidence is announced in Grand Trunk R. Co. v. Richardson, 91 U. S. 454. The saw-mill, etc., of Richardson, the plaintiff, was burned on the seventh of June, 1870. The evidence tended to show that the fire was communicated from one of two engines belonging to the company; the first, drawing a passenger train westerly, passing the mill about half past 1 o’clock in the afternoon; the other, drawing a freight train easterly, passing it about 4 o’clock the same afternoon. One. half to three fourths of an hour after the last-mentioned train passed by the mill, the fire was discovered burning on the westerly end of a covered railroad bridge, from which it was communicated to the saw-mill. The evidence of the plaintiff in error tended to show that the fire was not communicated by either of the engines complained of, but, on the contrary, from a constant fire at the end of the tramway, about one hundred sixty-three feet down the stream, on the same bank of the river, maintained at the westerly end of the railroad bridge for the purpose of burning edgings, stickings, slabs, and other waste material from the saw-mill. After the company *484had rested its case, Richardson was allowed to prove that at various times during the same summer,, before this fire occurred, some of the company’s locomotives in an unusual manner scattered fire in passing the mill and bridge, without showing either that those which it was claimed communicated the fire in question were among the number,, or that they were similar in their make, state of repair, or management to said locomotives. The engines were unknown and unidentified. Mr. Justice Steong-, in ruling upon this question, said: “The third assignment of error is that the plaintiff was allowed to prove, notwithstanding objection by the defendant, that at various times during the same summer, before the fire occurred, some of defendant’s locomotives scattered fire when coming past the mill and bridge, without showing that either of those which the plaintiff claimed communicated the fire was among the number, and without showing that the locomotives were similar in their make, their state of repair, or management to those claimed to have caused the fire complained of. The evidence was admitted after the defendant’s ease had closed. But whether it was strictly rebutting or not, if it tended to prove the plaintiff’s case its admission as rebutting was within the discretion of the court below, and not reviewable here. The. question, therefore, is whether it tended in any degree to show that the burning of the bridge, and the consequent destruction of the plaintiff’s property, were caused by any of defendant’s locomotives. The question has often been considered by the courts in this country and in England, and such evidence has, we think, been generally held admissible, as tending to prove the possibility and the consequent probability that some locomotive caused the fire, and as tending to show a negligent habit of the' officers and agents of the railroad company: ” citing Piggot v. Railroad Co., 3 Man. G. & S. 229; Sheldon v. Railroad Co., 14 N. Y. 218; Field v. Railroad Co., 32 N. Y. 339; Webb v. Railroad Co., 49 N. Y. 420; Cleaveland v. Railroad Co., 42 Vt. 449 ; Railroad Co. v. McClelland, 42 Ill. 358; Smith v. Railroad Co., 10 R. I. 22; Longabaugh v. Railroad Co., 9 Nev. 271.
In Sheldon v. Railroad Co., 14 N. Y. 218, the plaintiff gave evidence which tended to show that the engines used by the defendant lacked some apparatus which was in use upon some *485other locomotive engines, and which rendered the latter less liable to communicate fire to substances at the side of the road than those which were without that apparatus; that, shortly before the fire, sparks and fire had been thrown from the engines used by the defendant, in running its trains through the witness’s premises, a greater distance than this building stood from the track of the railroad; and that he had picked up from the track, after the passage of trains, lighted coals more than two inches in length. It was argued by the defendant’s counsel that the evidence was too remote and indefinite ; that it did not refer to any particular engine, etc. Chief Justice Denio, in delivering the opinion of the court, said: “ This argument is not without force, but at the same time I think it is met by the peculiar circumstances of this case. These engines run night and day, and with such speed that no particular note can be taken of them as they pass. Moreover, there is such a general resemblance among them that a stranger to the business cannot readily distinguish one from another. It will therefore generally happen that when the property of a person is set on fire by an engine, the owner, though he may be perfectly satisfied that it was caused by an engine and may be able to show facts sufficient, legitimately, to establish it, yet he may be utterly ignorant what particular engine did the mischief. It would be practically quite impossible, by any inquiries, to find out the offending engine, for a large proportion of those owned by the company are constantly in rapid motion. The business of running the trains on a railroad supposes a unity of management, and a general similarity in the fashion of the engines and the character of operation. I think, therefore, it is competent prima-facie evidence for a person, seeking to establish the responsibility of the company for a burning upon the track of the road, after refuting every other probable cause of the fire, to show that, about the time when it happened, the trains which the company was running past the location of the fire were so managed in respect to the furnaces as to be likely to set on fire objects not more remote than the property burned. It is presumed to be in the power of the company, which is intimately related with all its engineers and conductors, to controvert the fact sworn to if it is untrue, or, if true in a particular instance, that it was not so in respect to the *486engines which passed the place at a particular time before the occurrence of the fire. The effect of the evidence would only be to shift the onus probandi upon the company, and that, under the circumstances of this case, seems to me to be unavoidable.”
We may also refer to the case of Koontz v. Railway Co., (Or.) 43 Am. & Eng. Ry. C. 11, which was an action to recover damages for the destruction of plaintiff’s mill by fire falling from one of defendant’s locomotives. What particular engine this was the evidence did not disclose, nor was the plaintiff able to ascertain or make proof of its identification from other engines of the company; but, to strengthen the inference that the burning of the mill originated in sparks from this engine, and to show habitual negligence of the officers and agents of the railroad company, he introduced evidence to show that other engines, of like appearance and construction, frequently scattered fire in large quantities, and set other fires along the track, prior and subsequent to the burning complained of. Mr. Justice Lobd, in delivering the opinion of the court, said: “ On account of this difficulty of identifying a passing engine, especially at night-time, so as to make direct proof of such negligence, and also for the reason, as stated by Mr. Thompson, that the business of running railroad trains supposes a unity of management, and a general similarity in the construction of engines, the admission of evidence as to other and distinct fires from the one alleged to have caused the injury, is permitted. Nor is it requisite that the testimony must also show that the engine which it is claimed caused the fire was one of those which had previously or subsequently scattered fire along defendant’s track, but it is enough, as was shown, that it is similar in appearance and construction, and under the same general management. Hence it is quite generally held that evidence that sparks were frequently ejected from passing engines, causing fire along its track, on other occasions, is relevant and competent to show habitual negligence, and to strengthen and sustain the inference that the fire originated from the cause alleged. As the plaintiff must proceed with his evidence in the first instance, the fact that the defendant may be able to prove the identity of the engine, cannot have the effect to make the admission of such evidence error.” *487In Field v. Railroad Co., 32 N. Y. 339, the court, in speaking of this quality of evidence, says: “At all events, it showed that a practice was indulged in on the part of the company, about the time and near the place, which would have injured the plaintiff’s property, rendering it probable, to a certain degree, that the injury was attributable to that cause.”
We have quoted extensively from these authorities to show that the rule of evidence referred to, although, perhaps, comparatively new in its application in Pennsylvania, is the rule generally recognized in this country, not only by the text-writers, but by the courts. It may therefore be considered as settled, in cases of this kind, where the offending engine is not clearly or satisfactorily identified, that it is competent for the plaintiff to prove that the defendant’s locomotives generally, or many of them, at or about the time of the occurrence, threw sparks of unusual size and kindled numerous fires upon that part of their road, to sustain or strengthen the inference that the fire originated from the cause alleged. And as, in the case at bar, it is not definitely ascertained to which of the four engines this fire was attributable, three of them being unknown and unidentified, we cannot see how testimony of this character could be excluded.
But the objective point of the inquiry is the condition of the passing engines at the time of the occurrence. It is a matter of little consequence what may have been their condition ten years or two years before that; for the precautions against fire, and the management of the engines, may have been greatly changed within that period. It does not follow because the company, in its official management, may have been negligent in this respect at a time so remote, that it still remains so. The habits of individuals may, in some sense, be spoken of as fixed habits; but the official control and management of the affairs of a railroad company, as well as the various devices used as precautions against danger, are liable to frequent and radical changes. The line must be drawn somewhere. This class of testimony is exceptional in character at the best, and is only admissible because the ordinary sources of proof are inaccessible, and direct evidence impracticable. The rule should not, therefore, be carried beyond the necessity which justifies its admission. If at or about the time *488when fires are alleged to have been set by locomotive engines, unknown by number or other means of identification, the company is shown to have been habitually negligent in the equipment or management of its engines, or of many of them, this is a circumstance to be considered in connection with others, not only in determining the origin of the fire, but in deciding whether or not the company was, at the time, in this as in many other instances, negligent in failing to provide suitable precautions against danger. If many of the company’s engines, at or about the time, are without sufficient spark-arresters, and frequent fires are kindled in consequence, it may well be inferred, in-view of the effectual character of mechanical inventions of this kind, not only that the fire in question originated from this cause, but that it occurred from the habitual negligence of the company in failing to provide sufficient spark-arresters.
Reasonable latitude must, of course, be allowed. The purpose of such proofs would be defeated if they were confined to the exact or precise time of the occurrence. In Stranahan’s case, the court admitted proof of the extent to which the various locomotives of the company threw sparks on or about the ninth (sixth) of November, 1867, when the fire occurred. In Growen v. Glaser, the inquiry was as to sparks thrown and fires set very shortly before and very shortly after the occurrence. In Sheldon v. Railroad Co., supra, the inquiry was restricted to matters occurring about the time and near the place of the fire. In Koontz v. Railway Co., the offer was somewhat more extended in its effects, but we are of opinion that the rule should not be given greater latitude than we have given it.
In the case at bar, the first offer received, and which is the ground of the first specification of error, was as follows : “ Plaintiffs offer to prove that the property of persons along the line of defendant’s road, which passed the property of the plaintiffs destroyed by the fire in question on August 10,1888, and within twelve miles of plaintiffs’ said property, was repeatedly set on fire by unknown and unidentified engines of the defendant, and that the sparks causing said fires, emitted by the said engines, exceeded a hickory-nut in size: to be accompanied by evidence of experts showing that engines throwing sparks of the size of hickory-nuts either did not use the most ap*489proved spark-arresters in general use, or, if they did, the sparkarresters used were permitted to become defective and out of repair, or were negligently managed by those in charge of them.” This offer, it will be seen, was wholly without limit as to time. The testimony received under it was, in some instances, confined to two or three months, in some to six months, and in some the testimony was general, and in such form as not to indicate to what period of time it referred. The second offer was: “ To prove that many of the locomotive engines of the defendant, which they cannot identify and which passed the plaintiff’s mill frequently during a period of six months preceding the fire, habitually threw sparks of the size of a hickory-nut, or larger; ” etc. We are of opinion that the admission of these offers was error. The examination should be confined to the negligent operation of the engines of the company at or about the time of the fire, with such reasonable latitude, before and after the occurrence, as is sufficient to enable such proofs to be practicable.
What has been said disposes of the first, second, and third assignments of error. The remaining assignments are without merit, and are dismissed.
The judgment is reversed, and a venire facias de novo awarded.