Under our statute (§ 3785), the plaintiff *Page 79 was not bound to show that the fire in question was caused by the defendant's negligence, but only that it was set by one of the defendant's engines. In this case the plaintiff attempted to show that fact by circumstantial evidence, including evidence of the fact that the defendant's engines, as equipped and operated about that time and place, were liable to scatter sparks and set fires; and that its plant was within the range of cinders thrown from the defendant's engines. No doubt evidence of this kind is relevant and therefore admissible in cases where the particular engine charged with the fire is not identified. See 33 Cyc. 1371, citing cases from many jurisdictions. In this case the particular engine charged with setting the fire was identified, while the plaintiff was putting in his case, as engine No. 114, and the question is whether such identification narrows the issue to an inquiry into the habit or capacity of engine No. 114 in the matter of setting fires and scattering cinders, and makes irrelevant the evidence offered by the plaintiff as to the habit or capacity of defendant's engines not identified. The objection is that the argument from the general to the particular involves a non sequitur.
In those jurisdictions where the plaintiff's cause of action is founded upon the defendant's negligence, or upon a rebuttable presumption of negligence arising from the fact that the fire was set by a railroad engine, it seems to have been generally held, except in New York, that when the engine charged with setting the fire is identified, evidence of other fires not set by the same engine is inadmissible. Lesser Cotton Co. v. St.Louis, I. M. S. Ry. Co., 52 C.C.A. 95, 114 F. 133; Gibbons v. Wisconsin Valley R. Co., 58 Wis. 335,17 N.W. 132; Bell v. Chicago, B. Q. Ry. Co.,64 Iowa 321, 20 N.W. 456; Norfolk W. Ry. Co. v.Briggs, 103 Va. 105, 48 S.E. 521. *Page 80
We think the reasoning of those cases does not apply to this. In the first place, our statute eliminates the issue of negligence, so that the excluded evidence was not open to the objection of irrelevancy on the ground that it was offered for the purpose of proving that because the defendant was negligent in respect to some engines it was negligent in respect to No. 114; and in the second place, the finding contains a formal admission by the defendant that fire may be set from a locomotive engine although the spark arrester is in perfect repair. According to this record all engines are more or less liable to set fires, the difference between perfect and defective spark arresters being one of degree. Therefore the excluded evidence was not open to the objection of irrelevancy on the ground that its only purpose was to lead the jury to believe that because some engines were defective, No. 114 was also defective. It was relevant upon the issue of the probable cause of the fire, and the fact that engines with defective spark arresters are more liable to set fires than was No. 114, whose spark arrester was in good repair, goes to the weight of the evidence and not to its admissibility.
The supposed non sequitur is also bridged over by the evidence that engine No. 114 was seen scattering sparks before and after it passed the plaintiff's plant. This being so, the evidence of other fires about the same time and place supplies the major premise of the plaintiff's argument, which may be summarized as follows: engines which scatter sparks are liable to set fires; No. 114 scattered sparks; therefore, if other causes are excluded, it is probable that the fire was set by No. 114.
The same reasoning applies to the questions asked of the witnesses Sperry and Deane for the purpose of showing that the place where the plaintiff claims that *Page 81 the fire started was within range of cinders thrown from defendant's engines. When it is shown that No. 114 did throw out cinders, and admitted that engines with perfect spark arresters may do so, the fact that engine cinders were found on the roof of plaintiff's plant has some tendency to show that No. 114 may have set the fire at that distance from the railway track, although its spark arrester was in good repair; and so far as the objection of irrelevancy was concerned, the plaintiff was entitled to have the evidence go to the jury for what it was worth.
Evidence of other fires and of the emission of sparks, and of the finding of cinders, should be limited in point of time and place to facts having some probative value in establishing the probable cause of the particular fire in question; and it may be that some of the evidence in question might properly have been excluded on the ground that it was too remote or indefinite in point of time or place. This is a matter of judicial discretion, and we have not attempted to draw any distinctions upon that ground, because the record plainly shows that the trial court excluded all evidence of this kind on the ground of irrelevancy and not because it was too remote or indefinite.
That portion of the charge of the court which is set forth in the statement of facts was also erroneous, for the reasons given, in so far as it instructed the jury that the admitted fact that locomotive engines sometimes set fires had no bearing upon their deliberations.
The other assignments of error are not pursued on the brief.
There is error and a new trial is ordered.
In this opinion the other judges concurred.