The following opinion was filed October 5, 1909 :
Dodge, J.
1. The first error assigned is upon refusal to -submit to the jury the question whether the injury resulted “from voluntary exposure to unnecessary danger or obvious risk of injury.” According to the great weight of authority three elements are essential to this excuse from liability: (a) Conscious knowledge of the danger; (b) intentional or wilful exposure to it; and (c) that the danger shall be unnecessary. As to the first two elements,, this court has declared itself in accord with such authority in Schneider v. Providence L. Ins. Co. 24 Wis. 28; Shevlin v. Am. Mut. Acc. Asso. 94 Wis. 180, 68 N. W. 866; and in Sargent v. Central Acc. His. Co. 112 Wis. 29, 81 N. W. 796. The only evidence upon which the court or jury could act was that the deceased was a locomotive fireman insured as such, whose duties of course took him about .the tracks in railroad yards, and that on the *45occasiofi in question lie took an entirely usual route, and the shortest one, from his home to his place of employment at the roundhouse through the railroad yard and in the vicinity of tracks, and that his injuries indicated that he had been struck and run over by a passing engine. In the absence of any'other evidence, we agree with the trial court that a conclusion either that he knew of the danger from which he suffered, or that he wilfully and intentionally exposed himself to it, could have been based only on conjecture or guess. The burden of proof was upon the defendant to offer evidence from which such conclusion might result by reasonable inference and not alone by conjecture. Follis v. U. S. Mut. Acc. Asso. 94 Iowa, 435, 62 N. W. 807; Musbach v. Wis. C. Co. 108 Wis. 57, 84 N. W. 36; Hyer v. Janesville, 101 Wis. 371, 77 N. W. 729; Schell v. C. & N. W. R. Co. 134 Wis. 142, 113 N. W. 657. The trial court did not err in holding that no such evidence had been introduced, and therefore an affirmative answer to such question could not have been sustained.
2. A second error is assigned upon the refusal of the court to direct verdict that the deceased at the time of his injury was under the influence of intoxicating liquor, which fact it is claimed was established without dispute. We cannot at all agree with this view of appellant’s counsel. The evidence of any considerable use of intoxicating liquor, or influence thereof upon deceased even at any time during the evening before his death, is very conflicting; but, even if a condition of some degree of intoxication had existed, there was evidence tending to show that the last use of liquor was prior to 10 o’clock; that an hour or more of slumber succeeded it, and that as early as an hour before the casualty, on awakening from that slumber, deceased had fully recovered all his faculties and was free from any apparent influence of previous po-tations. We agree with the trial court that there was plenary evidence on which the jury might have based a negative answer to the question submitted to them.
*463. Error is assigned, upon instructions substantially to the ■effect that the phrase in the policy “under the influence of any intoxicant” meant not every and any influence however •slight, hut such degree of influence as would materially impair the deceased’s ability to care for himself and guard against casualties, and that such degree of influence was equivalent to intoxication in the ordinary meaning of that word; that the jury should not answer the question in the affirmative unless they found that he was “intoxicated” or “drunk.” The entire phrase of the policy in which these words occur is, “where the accidental injury is sustained-while the insured is insane, delirious, or under the influence ■of any intoxicant of narcotic.” The rule, of course, is tho-r■oughly established that in case of ambiguity the words of an insurance policy are to be read most favorably to the insured. Here the intimate association of the words “under the influ•ence of intoxicants” with the words “insane or delirious” at ■once suggests that the influence of intoxicants intended to be described has some similarity in character with insanity or •delirium. Again, it must be presumed that this provision is included in the policy for some practical purpose, and that therefore it is intended to describe a condition which at least might enhance or affect the insurer’s liability. The “influence of intoxicants” is a very elastic term. We are told by physicians and experimenters that the most trifling quantity of alcohol has some effect, and that its effect persists for days, if not permanently, so that one is literally under the influ•ence from a single ordinary portion. We know as a matter of common knowledge that one of the first influences may be to stimulate those very faculties of observation and alertness which would improve the capacity of the subject to shield himself from danger, or escape, and that some such degree of influence of an intoxicant would not in any respect increase the pei’il of injury. It is therefore a natural and almost necessary assumption that these words were not inserted in the *47policy for the purpose of depriving the assured of the benefit -thereof in case of every and any influence of intoxicating liquors, however slight and however nonprejudieial to the in.surer. The field, therefore, is open for construction to ascertain just what degree or kind of influence is referred to. As already said, we must presume that it means such and so much influence as impairs the ability of the subject to care for himself and thus increases the probability of his suffering accidental injury. In light of such reasoning it has been de-•cidedby all courts speaking upon the subject that influence of intoxicants in accident policies means the same thing as the word “intoxication.” 3 Joyce, Ins. § 2612; Standard L. & Acc. Ins. Co. v. Jones, 94 Ala. 434, 10 South. 530; Campbell v. Fidelity & Cas. Co. 109 Ky. 661, 60 S. W. 492; Jones v. U. S. Mut. Acc. Asso. 92 Iowa, 652, 654, 61 N. W. 485; Prader v. Nat. M. Acc. Asso. 95 Iowa, 149, 63 N. W. 601. In this field of indefiniteness it- is important that some exact line should be adopted by which the rights of parties are rendered certain, and, in absenee_of any cogent reasons to the contrary, we deem it wise and justifiable to adopt this line of demarcation which has been approved by express decision of other courts; especially since such decisions preceded the date of this contract and may reasonably be presumed to have been In mind when its phraseology was adopted. We conclude that the instruction was substantially correct.
: 4. Refusal of certain requested instructions is assigned as ■error. They assumed a state of intoxication at some time during the evening prior to the death of insured. Since, ■as we have said, this fact was in dispute, the requests were improper in form and no error was involved in their refusal.
By the Court. — Judgment affirmed. ■
Wiusnow, C. J., took no part.
A motion for a rehearing was denied December 7, 1909.