The plaintiff introduced, without objection, evidence tending to show that defendant had permitted dry grass to remain between the track of the railway and the fence, in such quantities as would constitute a means of communicating fire from the engine to the field beyond the fence. The defendant subsequently moved to strike out all testimony relating to the condition of the strip of land between the track and the fence, “for the reason that it appeared from the whole evidence introduced by the plaintiff that the combustible material, if any, on said strip was not an element in causing the ignition or burning of plaintiff’s property.” The denial of the motion to strike out by the court below is assigned as error.
The motion was not based upon the ground that the testimony was inadmissible under the averments of the complaint, and that objection cannot first be made here. 'Where a motion is made to strike out testimony the moving party should specify his objections to the testimony with like particularity as is required in pointing out an objection to a question. (Sill v. Reese, 47 Cal. 341.) If the dry herbage was permitted to remain standing in such quantities as showed negligence, evidence of that fact, in the absence of an objection of variance from the allegations of the pleading, would ordinarily be admissible. (Flynn v. S. F. & S. J. R. R. Co., 40 Cal. 14.)
There was some evidence that the grass on defendant’s “ right of way” had been recently burned, and it was a question of fact for the jury whether the fire was lighted there or in the field adjacent. If, however, as argued by counsel for appellant, the evidence was conclusive that the fire commenced in the field of Cagney, we ought not to assume that *182the jury, without any evidence, found that it commenced on the “right of way.” If the court had granted defendant’s motion, it could not have obliterated from the memories of the jury the evidence already received. The order granting the motion would have constituted an implied direction to the jury to disregard the testimony. In the charge of the court the jury were distinctly told that if the fire was not communicated by the dry grass or herbage on the line of the railroad, the condition of the “right of way” as to dry grass or herbage was immaterial, and required no consideration at their hands. It must be presumed, in view of the charge, that no injury was done defendant by the denial of his motion to strike out.
The court below properly refused a nonsuit. We think there was evidence tending to prove that the fire was not the probable result of the ordinary working of a locomotive under like circumstances, and in such case evidence that the fire was communicated from the engine is evidence of negligence sufficient to go to the jury. (Hall v. Sac. V. R. R. Co., 14 Cal. 387.) There was, however, evidence of specific negligence, in that there was evidence tending to prove that the particular engine was required to perform service which caused it to “ labor,” and to emit more sparks than if a less( number of cars had been attached to it.
Again: It is said that the nonsuit should have been granted, inasmuch as the fire was not kindled in plaintiff’s field, but in the field of one Cagney, an adjoining proprietor, from which it extended into the field of the plaintiff.
The legal proposition involved in the foregoing statement is, that if by negligence a fire shall commence on the premises of one proprietor and spread from thence to those of another, the latter shall never have his action against him guilty of the negligence. We think this proposition cannot be maintained. To refute it, it is not necessary to establish the counter proposition that - the adjoining proprietor thus injured shall always recover. It may be assumed, perhaps, that a city fire which has its origin in one building will not ordinarily extend throughout a block, and yet a jury may be justified in saying, when a fire is started in a field which con*183stitutes a portion of a larger tract of dry grass, or corn fully ripe, that it will usually be driven into another field, from which the first is separated only by a fence of boards. It is a rule applicable to all cases of mere negligence, that the wrongdoer is liable for the proximate and not for the remote consequences of his default. A long series of judicial decisions has defined proximate, or immediate and direct damages to be the ordinary and natural results of the negligence; such as are usual and as, therefore, might have been expected; and this includes in the category of remote damages such as are the result of an accidental or unusual combination of circumstances, which would not be reasonably anticipated, and over which the negligent party has no control. In Ryan v. N. Y. C. R. Co., cited by appellant’s counsel, the court held when a railroad company negligently set fire to its own woodshed and the sparks communicated the fire to and destroyed the house of another proprietor, one hundred and thirty feet from the shed, that no cause of action existed in favor of the owner of the house. (35 N. Y. 210.) In that case no new principle was involved. The learned judge who delivered the opinion placed the decision on the ground that the burning of the house was not to be expected from the first firing. (See Webb v. R. W. & O. R. Co., 49 N. Y. 420.) It would be strange if, among the numerous cases in which resort has been had to the rule of the law of negligence to which we have referred, courts had not differed in their application of the rule; but for the purposes of this action it may be admitted that the rule was properly applied in Eyan’s case. We are still confident, considering the long dry season of California and the prevalence of certain winds in our valleys, that it may be left to a jury to determine whether the spreading of a fire from one field to another is not the natural, direct or proximate consequence of the original firing.
W e think there was no error in permitting proof that prior and subsequent to the fire which produced the injury complained of, other fires were kindled by defendant’s engine. The evidence was confined to fires caused by the same engine, in the same vicinity and about the same time.
*184Similar evidence, except that no time was specified, was said by the Court of Appeals in Maryland to have been properly rejected. (Balt. & Susq. R. Co. v. Woodruff, 4 Md. 254.) And the Supreme Court of Pennsylvania seems to have held that the firing of a plaintiff’s woods, on several other occasions, by a defendant’s locomotives, was not evidence from which could be inferred negligence in the management of a particular engine at a certain time. (R. R. Co. v. Yeiser, 8 Pa. St. 366.) On the other hand, ithas been determined in New York that evidence of the frequent dropping of sparks and burning coals by engines passing upon the same road on previous occasions is relevant and competent to show negligence, and to make it probable that plaintiff’s injury proceeded from the same quarter. (Field v. N. Y. C. R. Co., 32 N. Y. 339; Sheldon v. Hudson River R. Co., 14 N. Y. 218.) This is’the view adopted by the learned authors of an exhaustive work on the law of negligence. (Shearman and Bedfield, 333.) And such seems to be the law as declared by the Supreme Court of Iowa. (Gandy v. C. & Northwestern R. Co., 30 Iowa, 422.)
We think the evidence objected to tended to prove that the fire in question was caused by sparks from the engine, and also that there was something wrong in the management or defective in the construction of the engine. Evidence is admissible if it is of a circumstance which, with other circumstances, may bring home to the mind a conviction of the main matter in issue. There may be difficulty in many instances in drawing the line of separation between evidence which thus tends to establish the issue, or from which, in accordance with ordinary experience, a reasonable inference may be drawn as to the principal matter in dispute, and evidence of facts simply collateral. The last class of evidence is not admissible. (1 Greenl. Ev. 448.) In the case before us, it may be possible that defendant was not prepared to produce evidence (which in reality existed), tending to prove that the other fires were not caused by its negligence. There is nothing in the record, however, to suggest such surprise; and when evidence tends to establish the ultimate fact, the inconvenience *185of disproving the circumstance testified to, or of showing that the usual and natural presumption to be drawn from the circumstance ought-not to be indulged in, does not always constitute a reason for rejecting the evidence. The admission or rejection of such evidence may resolve itself into a question of policy and expediency dependent upon the pressing necessity of resorting to circumstantial evidence in cases in which direct evidence is not ordinarily to be had.
We find no material error in the instructions. The bill of exceptions contains no specifications of the particulars wherein the evidence fails to sustain the verdict.
Judgment and order affirmed.
Mr. Chief Justice Wallace did not express an opinion.