WOLVERTON, District Judge.
This is an action by the defendant in error to recover damages for the death of her husband, Samuel Blum, under a policy insuring against accident issued by the plaintiff in error. The policy covers “loss or disability resulting from bodily injuries effected directly and independently of all other cause through external, violent and accidental means (excluding suicide, sane or insane, or any attempt thereat).”
At the close of plaintiff’s case a motion for nonsuit on the part of the defendant was denied. At the close of the entire evidence the defendant demurred ore tenus to the sufficiency of the evidence to carry the case to the jury and in the alternative moved the court for a directed verdict in its favor. Both were denied by the court. The verdict was for plaintiff.
Thé assignments of error are three:
First, that the court erred in denying defendant’s motion for non-suit.
Second, in overruling the demurrer to the evidence.
Third, in refusing to instruct a verdict for the defendant.
All for reasons: (1) That the complaint fails to state a cause of action; (2) that there was a total failure of proof of death from accidental causes; (3) that there was a total failure of proof of any death effected directly and independently of all other causes through external, violent, and accidental means; and (4) that there was a total failure of proof of death caused in any manner covered by the provisions of the policy of insurance.
There was no demurrer to the complaint; nor were there any exceptions reserved to any instructions given by the court or any-refusal of the court to give requested instructions.
The complaint simply alleges, so far as material here:
That on January 12, 1917, “the said Samuel Blum died in the city of Seattle, county of King, state of Washington. His death was effected through external, violent, and accidental means.”
The answer denies the latter clause, and for an affirmative defense alleges:
[949]*949“That the policy relied upon by the plaintiff expressly excludes from the risks assured death effected through-suicide, sane or insane, or any attempt thereat, and the death of the said Blum was effected through suicide or an attempt thereat, and not otherwise.”
Thus are presented the particular issues upon which the case was tried and went to the jury.
[1] The suggested insufficiency of the complaint is not insisted upon in the briefs; nor was it at the argument. Even if it had been, the complaint must be held to be sufficient after verdict, when all intend-ments operate in its support.
Nor is the demurrer to the evidence insisted upon. There is therefore but one assignment of error that calls for notice, namely, the one relating to the refusal of the court to direct a verdict for the defendant. Thus is presented but the one question whether the testimony adduced at the trial was sufficient to warrant the court in sending the case to the jury. This requires a statement of the salient facts as disclosed by the evidence.
Blum was past 40, had been in business for a number of years, and was prosperous, having acquired an estate rated at $125,000. With this he was satisfied, as it was remunerative, and he was not eager to accumulate greater wealth. He was happy and contented in his domestic relations, and was esteemed and trusted in his social and business relations. On January 1, 1919, certain of his property in Alaska was burned, news of which came to him on the 2d. This greatly depressed and worried him at the time, and he complained of loss of sleep, and that he could not get the Alaska fire out of his mind. His condition of mind improved somewhat, however, prior to January 12th, the date of his death.
As bearing upon this, and as respects his disappearance from his office when last seen therein, Laura Lawson testified, among other things, that she was a public stenographer and had a desk in Blum’s office, but in a room separate from his, with a door communicating, and that she did his stenographic work; that Blum was in the office on the morning of the 12th when the witness came in about half past 9; that he asked hér, as was his habit, how he looked, to which she replied, “You look like you had a good night’s sleep.” He smiled, but remarked, “Well, I slept sound, but I was full of dope,” or something of that order. He was worrying, and his head ached all the time from worry and nervousness. He was told not to worry, that there was nothing to worry about, and he said, “That is all they tell me,” and wished he could see it that way. She further testified that the worry started with the fire at Valdez; that he was like anybody else when there is something on his mind, and walked forth and back, and would talk about it. He received telegrams from time to time about the fire, and when he found one thing to be all right, then he would wonder if something else was all right, and kept saying if he had been up there he would feel differently, but because he was down here naturally he imagined a whole lot was really worse than it was. He manifested his worry by walking forth and back, and putting his hand up to his head, and would say “so much all during the day” that he [950]*950could not collect his thoughts. Witness never heard him say that before the 1st of January. She saw no outward indication of any change relative to his being worried between the 2d and 12th of January, and he complained that he could not sleep. Mrs. Blum was with him in the office in the morning of the 12th, and they went out for an automobile ride. As he left he said he was going up to the house to take a nap, and would be back some time during the afternoon. He came back around half past 2 or 3 o’clock, walked into his office and came out again, and walked forth and back in the two rooms, and, probably half an hour later, said he was going to get shaved, remarking, “We have to keep clean anyhow, no matter what happens.” He was gone an hour, or a little longer, and came into the office again, and walked forth and back. Witness was busy at the time he went out again. As he passed her desk, he said he was going for a little walk, and would be back in a few minutes. He went out into the hall; then came back into the office. He then started pacing forth and back in both rooms, and did that for probably 10 or 15 minutes, and then' went into his own office, half closing the door as he went. Very shortly, in a minute or a minute and a half, witness missed him from his office, and on going in found he had disappeared. Then it was ascertained that he had fallen from the window in the room.to the sidewalk below. Witness testified that Blum showed no indication of faintness on the day that he died, but it was developed on cross-examination that he had a fainting spell in the office several days previous thereto, but was revived shortly by a stimulant. This testimony has corroboration.
The barber, Mr. Hurd, relates that Blum came into his place about 4 o’clock, took off his coat and hat, got a paper from the porter, and walked back and forth across the floor until a chair was ready for him, which was about 20 minutes. While shaving him, witness asked if he had been sick, to which he replied “No,” but said that he had not slept for two weeks. When asked what was the matter, he said, “That fire — I can’t get it off my mind,” and further, “I just feel awful; I never felt this way in my life before; I feel terrible.”
Mrs. Estelle R. Blum, wife of deceased, testified that their married life had always been happy, and that Blum was kindly towards every one; that he had a habit, when engaged in thinking or discussing business or otherwise, of pacing the floor, which he did at home as well as in the office. The news of the fire at Valdez came on January 2d, while they were at breakfast, and Blum went down at once to ascertain if there were further news. Witness could not recall any special time of his being dizzy, but remembered that he did complain of dizziness; that he would often want the fresh air, and when he did he would throw up the window or open the front door, and at times would go outside. He showed a worried condition during the period following the news of the fire. The news came very slowly. He worried about the insurance, or rather as to the responsibility of the companies in which he was insured. When assured of their responsibility, he was very much relieved. Then he was in doubt whether he would be allowed to continue in his mercantile business while [951]*951awaiting fire adjustment, but was assured later that he would, and that worry was disposed of. Again, he worried over the uncertainty of the safety of his papers in the vault, but was relieved when he had news that they were all right. After that he still continued to be perturbed about the health of his cousin Meyer Blum, who was associated with him in business in Alaska, and that worried him somewhat up to the time of his death. The conditions in Valdez were a constant topic of their conversation. While the other things had cleared up, he continued to worry somewhat about Mr. Meyer Blum’s health. He always worried about his cousin’s health, both before and after the fire. After the fire he was very, very wakeful, but each night he obtained more sleep. He never had been a good sleeper, said the witness; “he had never slept a night through, all our married life.” After the fire he slept better each night than the preceding, and each day was an improvement over the preceding day. About 8:30 on the morning of the 12th witness and her husband left the house in an automobile, and drove down to the boat landing, to see Mr. Ford, the insurance adjuster, off for Valdez. Blum was in very good spirits, and appeared well physically, and “he seemed his natural self.” His conversation at the dock with Ford was jovial, and he interested himself with the purser of the boat for Ford’s convenience on the voyage. From the dock they went to Blum’s office, where they remained for perhaps half an hour. In the meanwhile Blum was walking up and down, and the conversation was general. From the office the two went for a ride, and eventually drove home. Blum complained that his feet were cold, and that he was sleepy, and he lay down on the bed. Witness made him comfortable, and he fell asleep. He slept about an hour and a half, and they had lunch together.- He lit a cigar, and remarked, “I feel fine; I feel better than I have for days,” and departed for his office. He was averse to taking medicine to induce sleep.
Blum is described as a man 5 feet 7 inches in height and weighing 190 pounds. The stool or window seat of the window from which he escaped was 39% inches above the floor. The window seat is 12% inches in width. From this there is a drop of 1% inches, with the wood sill sloping an inch; then another drop of 2% inches to the surface of a stone sill 17% inches in width, which slopes one inch. There is then a further drop of 4% inches to a stone belt course, which runs around the building and projects an additional 9 inches, with a half-inch slope; the entire width of the coping, including the window sill, being 43 % inches, and the drop on the whole 10 inches. The window opening, when the lower sash is raised to its limit, as it was found after Blum disappeared, comprises a space of 28% inches in width by 33% inches in height. An ordinary office chair was found standing near, with its back to the window.
Two witnesses observed the deceased while falling to the sidewalk —one when he was a few feet above the walk, and the other apparently just as he left the window ledge. He encountered a guy wire of the electric lighting system in his fall, and the globe of the arc light was broken. They describe him as falling limp, with his arms hanging [952]*952lower than his head, but not directly over the head like a person diving.
At the trial, when it was shown that death- had ensued from external and violent means, it was agreed between counsel that plaintiff had made out a prima facie case. Thereupon the defendant proceeded with its case. A question has arisen as to which of the parties had the burden of proof.
[2] We think it indisputable that, while the plaintiff made out a prima facie case by showing death by external and violent means that the defendant was required to overcome, yet in the end it was incumbent upon the plaintiff to establish by a preponderance of the evidence the fact as alleged by her that death was effected through external, violent, and accidental means. This is a burden which-the plaintiff could not escape.
Suicide is the antithesis of death by accidental means, and, of course, if it appears that Blum took his own life, plaintiff cannot recover under the policy. But this does not shift the burden. The clause in the policy requiring that death shall result through accidental means independent of all other causes devolves the burden upon plaintiff that she substantiate this also before recovery can be had. Travelers' Insurance Co. v. McConkey, 127 U. S. 661, 666, 8 Sup. Ct. 1360, 32 L. Ed. 308; National Masonic Act. Ass’n v. Shryock, 73 Fed. 774, 20 C. C. A. 3.
The deceased came to his death by one of three means. He either died through natural causes (that is, by sudden demise) and fell from the window, or he voluntarily threw himself therefrom, or he fell from the window or the coping outside through accidental means.
[3, 4] There is a presumption that death was not suicidal. This is disputable, but to overcome it, there must be proof of a tendency to establish the fact sufficient for the jury’s consideration. Without else, there is no presumption that death ensued through accidental means. The presumption is rather that it was the result of natural causes. Laessig v. Travelers’ Protective Ass’n, 169 Mo. 272, 280, 69 S. W. 469. So we get back to the burden that plaintiff must sustain, which is to show that death came through accidental means, and this independently of all other causes, with the presumption of death through natural cause to overcome, and that it was not suicidal in her favor. Was there evidence under this hypothesis sufficient to carry the case to the jury?
[5] It goes without saying that, in order for plaintiff to recover, there must be evidence that an accident occurred conducing to the injury. This does not mean, however, that there must be eyewitnesses to the accident or direct proof of the pertinent fact. The fact is susceptible of proof, as any other given fact, and it may properly be deducible by inference and presumption from facts proven; that is, the fact of accident may be established by circumstantial evidence, as other pertinent facts may be established under the rules of evidence. Brunswick v. Standard Acc. Ins. Co., 278 Mo. 154, 213 S. W. 45, 7 A. L. R. 1213.
[6] Much speculation is indulged in as to how Blum got out of the window. It is obvious, however, that a man of his size and activity by [953]*953stooping could have readily stepped from the chair, which it may be that he did, across the window seat to- the coping outside, and could have sat down on the window seal or safely stood on’the coping to get the air, if that were his purpose. The stone sill under the window seat was 1714' inches in width, with a slope of but an inch, while extending beyond that and 4% inches lower, was the belt course 9 inches in width, with a half-inch slope. One docs not have to enter the realm of conjecture to affirm that a man of Blum’s plij^sique, with his senses about him and his ordinary activity in movement, could safely have adjusted himself outside the window so that he would not have fallen in getting the air, if such were his purpose. It is also a reasonable hypothesis that Blum fainted while in the act of stepping through the open window, and in fainting pitched forward with such momentum as to cause his body to slide over the coping and be precipitated to the street below. These, with other hypotheses as well, might reasonably account for his getting out of the window without suicidal intent.
It is obvious from these observations that the circumstances and conditions present as shown by the evidence do not repel all reasonable hypotheses of accidental death or death by accidental means.
[7] When a man is found dead under circumstances showing clearly that he killed himself, there can be no presumption that death was accidental (Supreme Tent. K. of M. of the World v. King, 142 Fed. 678, 73 C. C. A. 668). or that it was from natural causes, for that matter. Johns v. Northwestern Mutual Relief Ass’n, 90 Wis. 332, 63 N. W. 276, 41 L. R. A. 587, is an apt illustration of the postulate. There the assured was found dead in a cistern. The aperture through which entry took place was so small that it could not, by any reasonable hypothesis, be inferred that he fell in while casually walking or passing over the cistern. So it was held that the inference of self-destruction was so strong as to overcome any presumption of accident that might otherwise be indulged.
Such is not the case at bar, for it is fully apparent that death might have occurred through other causes, or by other means. The question, then, whether death was suicidal was one among others to be submitted to the jury.
Counsel for plaintiff in error suggests a hypothesis which supposes that the- deceased, not having been affected with dizziness, had voluntarily placed himself head foremost at the outer ledge of the window, and so near the outer edge as to be within the danger zone and beyond the point where the laws of gravity and friction would tend to hold him; that is, in a position where the laws of gravity would tend to pull him over the ledge to the sidewalk beneath. If such were the case, it is argued that, although the fall might have been involuntary, the means which he selected would be voluntary, and the injury following could not be held to have been occasioned by external, violent, and accidental means. Another hypothesis of a similar nature is stated, but the one noted is sufficient for intelligent discussion of the legal proposition invoked.
[8] A distinction is sought, to be made between an accidental death and death through accidental means. The distinction seems to be borne [954]*954out by authority. Olinsky v. Railway Mail Ass’n (Cal.) 189 Pac. 835. Accidental death is aptly illustrated by the case of Shanberg v. Fidelity & Casualty Co. (C. C.) 143 Fed. 651, where the assured, while engaged in carrying a door, fell down and expired. An autopsy disclosed that he had a ruptured heart. It was held that death did not-result from “extraordinary, violent, and accidental means, independent of all other causes,” within the meaning of the policy, and that the rupture of the heart was the result, and not the means through which it was effected.
Another case (Martin v. Interstate Business Men’s Acc. Ass’n [Iowa] 174 N. W. 577), is where the insured ate oranges, engendering gastritis which caused death. It was held that the act of selecting the oranges was voluntary, and that the consequences, although unexpected or unintentional, could not be said to have resulted through “accidental means.”
In further illustration of the distinction, the Supreme Court has apparently approved an instruction:
“That, if a result is such as follows from ordinary means, voluntarily employed, in a not unusual or unexpected way, it cannot be called a result effected by accidental means; but that if, in the act which precedes the injury, something unforeseen, unexpected, unusual occurs which produces the-injury, then the injury has resulted through accidental means.”
This instruction was given in a case arising from the circumstance of the injured person jumping from a platform. Two other persons jumped from the platform at the same time and sustained no injury, and the case turned upon whether there was anything accidental, unforeseen, involuntary, or unexpected in the act of jumping from the time the deceased left the platform until he alighted upon the ground. A verdict for the beneficiary was upheld. Mutual Accident Association v. Barry, 131 U. S. 100, 121, 9 Sup. Ct. 755, 33 L. Ed. 60.
Now, extending the doctrine of this latter case to what might have-happened with Blum, if he deliberately placed himself in a position where he would fall, realizing that such would be the result of his act, then it could not be said that his injury came about through external, violent, and accidental means within the terms of the policy; this because the act would be voluntary and deliberate, and the result could not have been unforeseen. But if he voluntarily placed himself in a position of danger, even recklessly though it may have been, and through accident, by misstep or miscalculation, or through the force of gravity not anticipated or foreseen, and was precipitated from the window ledge, the cause of the injury would have been through accidental means within the meaning of the policy.
After all, however, these are hypotheses that went to the jury under the evidence, and we must assume under proper and applicable instructions, as no exceptions are here to any instructions given or to any requested and not given.
Again, there is brought into the controversy the supposition that Blum’s condition of health may have contributed directly or indirectly to his injury; it being argued that, where disease or physical infirmity concurs in producing death, the risk is not covered, because, while [955]*955death may have resulted from accident, it would have to be shown that the accident was the sole cause of death. Be this as it may, we must again assume, under the state of the record, that the jury was properly instructed in the premises and that they gave attention to the evidence in view of the rule of law applicable. But more of this later.
Another feature of the controversy is that Blum might have jumped from the window while in a condition of mental irresponsibility; that is, while in a state of melancholy, when his impulse was to take his own life.
Upon this, question medical experts were called, with the result of a divided and discordant opinion. Thus it became a matter for the jury to determine, and their finding respecting it was comprised by the general verdict.
Within the range and scope of the testimony, certain experts in physics gave evidence touching the possibility of the deceased’s falling out of the window, supposing him to have been in certain defined positions in relation to it, and including also the chair that was found by the window immediately after his disappearance.
This testimony is interesting and highly instructive, as bearing upon the hypotheses concerning which their opinions were desired. But there is an obvious infirfnity attending it, in that it did not cover other reasonable hypotheses respecting positions in which the deceased might have been immediately prior to his falling from the window or from the ledge outside.
Coming again to the one question involved, whether the evidence was sufficient to carry the case to the jury, we will give attention to some of the different phases of the evidence and the inferences that the jury was at liberty to draw therefrom.
[9] Blum had been a man in comparatively good health up to the time he received news of the fire in Valdez. He had been afflicted with dizziness on occasions, which induced him to seek the open air, from which he invariably obtained relief. Further than this, he had not slept well in later years. But these were matters to which little attention was paid, and which were not vital so far as his general health was concerned. After the news reached him from Valdez, he became greatly perturbed and worried, complaining of his head and of being unable to collect his thoughts, and at one time, a few days before his death, he had a fainting spell in the office. Aside from this, there were indications that he was restless, and felt bad, even as he had never felt in his life before. This evidence formed the basis of an inquiry as to whether he was not suffering from an attack of agitated melancholia, and whether or not it led to the taking of his own life. This was disposed of, and properly so, by leaving the matter to the jury for its determination. The evidence was such that a reasonable inference might have been drawn either way, and was unquestionably sufficient to require submission to the jury.
As we have seen, the presumption is that he did not commit suicide, and, further, the testimony in the case is not of such a nature as to exclude all reasonable hypotheses of death through accidental means. So the question of suicide, sane or insane, was properly one for the [956]*956jury. Pythias Knights’ Supreme Lodge v. Beck, 181 U. S. 49, 53, 21 Sup. Ct. 532, 45 L. Ed. 741.
There being testimony in the record sufficient for the jury’s consideration as to whether death resulted from suicide, we- may proceed to another phase of the inquiry; that is, whether death might have resulted from natural cause, as by apoplexy, heart failure, or the like.
[10] Here, again, Blum’s state of health was a matter for consideration, along with the fact that he in some wajr got out of the window, whether by struggle after death had stricken him, or by his own volition while retaining his mental and physical activities.
It is possible that he may have gotten out of the window while in an effort to get the air, and then have been stricken and life become practically extinct before he fell to the sidewalk. There is some indication that such might have been the case. An eyewitness saw the body almost at the instant it left the window ledge, and he says it appeared to be limp with the arms hanging downward. But this is only one of the many circumstances to be taken into account. While such a thing was possible, it was within the province of the jury to inquire whether it was a probable inference to be drawn, under all the circumstances surrounding his death. required the exercise of physical effort to get out of the window. This he was fully able to put forth when he was last seen, but a minute or two before he disappeared. So the postulate that he dropped dead after getting on to the window ledge is attended with questionable probability.
[11] There is this as well to be considered — which was also for the jury — that the fall itself was sufficient to produce death, and the condition in which the deceased’s body was found was some indication at least that death resulted therefrom, and of course by violent and external means.
“That the courts will presume,” says the court in Jenkin v. Pacific Hut. Life Ins. Co., 131 Cal. 121, 124, 63 Pac. 180, 181, “that the death was the result of an accident, when nothing more is shown than that it was brought about by a violent injury, and the character of such injury is consistent with the theory of accident, seems to be a rule upheld by the great weight of authority.”
Again the court says, in Nichols v. Commercial Travelers’ Eastern Accident Ass’n, 221 Mass. 540, 543, 109 N. E. 449, 450, a case in which the deceased was ejected from a sleeping car berth while the car was in motion, either through his own volition or involuntarily, through a screen in the window comprising a space of from 24 to 25 inches in width by 18 inches in height, and was found dead under circumstances indicating that death was caused by a fall from the car:
“When a man is found dead under such circumstances, when the marks on the body show the cause of death and all the circumstances exclude the theory of disease or injuries intentionally inflicted by a third person, it may be inferred that the death was accidental and was not the resttlt of intention or design.”
The principle involved is concretely stated in 1 C. J. 495, as follows :
[957]*957“The fact of death does not of itself create any presumption that it was the reunir. of an aceident; and where, in order to make out plaintiff’s ease, it is necessary to base a presumption that death resulted from an injury on a presumption that the insured sustained an accidental injury, no recovery can be had. Where, however, it is apparent that the injury to or death of the insured was the result of external and violent means, and the issue is as to whether it was due to an accident, within the meaning of the policy, or to some cause excepted by the policy, the presumption is in favor of accident and against the existence of facts bringing the case within any of the exceptions of the policy, such as insanity of the insured, intentional injury inflicted by a third person, lack of due care and diligence, self-inflicted injuries, and suicide. These presumptions may, however, be overcome by facts and circumstances establishing the contrary.”
The authorities are generally to the same purpose. See Buckley, Adm’x, v. Massachusetts Bonding & Insurance Co. (Wash.) 192 Pac. 924, where a number are reviewed.
[12] Recurring again to the suggestion that death may have been the result of concurring causes, that is, of disease and accident, and therefore not within the terms of the policy, because of the language covering death “effected directly and independently of all other causes,” under such conditions the inquiry simply resolves itself into one of proximate cause.
“A sick man,” says the court in Bohaker v. Travelers’ Insurance Co., 215 Mass. 32, 34, 102 N. E. 342, 314 (40 L. R. A. [N. S.] 543), “may be the subject of an accident, wlilc-h but for his sickness would not have befallen him. One may meet his death by falling into imminent danger in a faint or in an attack of epilepsy. But such an event commonly has been held to be the result of accident rather than of disease.”
In the same case it is observed that:
“The language of this contract to the effect that the ‘accidental means’ must have operated ‘independently of all oilier causes’ to produce the death does not change the general rule of lavs that the proximate, and not a remote,.cause is the. one to which the law looks.”
Our attention has not been called to any clause in the policy that precludes recovery on the ground that disease has operated concurrently with “accidental means” to produce injury or death. The coprt is not required to search beyond the active, efficient procuring causé, to the cause of a cause, so that, as the court further observes:
“When one single predominant agency is disclosed, directly producing as a* natural and probable result the injury, which is accidental, and which operates independently of other like causes, then the effectual means required by the policy have been found.”
The language of Judge Taft in Manufacturers’ Accident Indemnity Co. v. Dorgan, 58 Fed. 945, 954, 7 C. C. A. 581, 590 (22 L. R. A. 620), is illustrative and highly instructive. He says:
“if the deceased suffered death by drowning, no matter what was the cause of his falling into the water, whether disease or a slipping, the drowning, in such ease, would be the proximate and sole cause of the disability or death, unless it appeared that death would have been the result, even had there been no water at hand to fall into. The disease would be but the condition; the drowning would be the moving', sole, and proximate cause.”
[958]*958So to apply the principle here, unless it can be said that Blum’s disease, if it can be so termed, or if he was so afflicted — about which there is a serious question under the evidence — was the proximate cause of his injury and death, then we must look elsewhere for the active and efficient cause. It is obvious that, under the evidence, it cannot be said, as matter of law, that disease was the proximate and efficient cause of Blum’s death. The case, as we have previously observed, was one, therefore, for the jury’s determination as respects the real and producing cause of death.
[13] True, the jury cannot be permitted to find its verdict upon conjecture and surmise; but, from a careful survey of the entire testimony found in the record,' we are assured that there is afforded a much more stable basis for inference and deduction, and that it was quite sufficient whereon to submit the case to the verdict of the jury.
We find no error in denying the motion for a directed verdict for the defendant.
Affirmed.