Andrew McAllen, the surviving husband of Effie W. McAllen, deceased, appeals from a decree of final distribution in the Matter of the Estate of Ellen M. Wallace, Deceased, whereby the latter’s estate was distributed to the devisees and legatees of Martha A. Allison, deceased.

Ellen M. Wallace was the mother of Effie W. McAllen. The two perished in the same calamity. Martha A. Allison, now deceased, was the mother of Ellen M, Wallace. She died subsequently to the tragedy in which her daughter and granddaughter were killed. Ellen M. Wallace left a last will whereby she bequeathed and devised all of her property to her daughter Effie, who was -her only child. If the latter survived her mother, even for an instant, she would take under her mother’s will, and in that case the appellant here, as the surviving husband and only heir at law of Effie W. McAllen, would be entitled to have distributed to him the estate which was left by Mrs. Wallace and which would have gone to Effie McAllen had the latter lived to receive it. If, however, Effie was survived by her mother, though but for an infinitesimal interval of time, then the bequests and devises which Mrs. Wallace made to Effie lapsed, and in that event the grandmother, Martha A. Allison, as Mrs. Wallace’s sole heir, would inherit all of the latter’s estate. The trial court found that Mrs. Wallace did survive her daughter Effie, and accordingly distributed her estate to the devisees and legatees of Martha A. Allison, who, as we have stated, died subsequently to the calamity in which Mrs. Wallace and her daughter lost their lives. The sole question therefore is: Do the facts warrant the finding that Mrs. Ellen M. Wallace survived her daughter Effie W. McAllen 1

Ellen M. Wallace, Effie W. McAllen, and three other persons were killed in a collision between a railroad train and the automobile in which they were riding. Without doubt the disaster was a “calamity” within the meaning of the statute with reference to presumptions of survivorship where persons are killed in the same calamity. (Code Civ. Proc., sec. 1963, subd. 40.) At the date of the catastrophe Mrs. Wallace was over sixty and her daughter was over fifteen but less than sixty, Mrs. Wallace’s age being sixty-eight while her daughter was of the age of fifty. Appellant contends that is is impossible to determine from the evidence with any satisfactory degree of accuracy whether Mrs. Wallace or Mrs. McAllen died first, and that therefore the presumption of the statute should prevail. Subdivision 40 of section 1963 of the Code of Civil Procedure, so far as it is applicable to the facts of this case, reads: “When two persons perish in the same calamity, such as a wreck, a battle or a conflagration, and it is not shown who died first, and there are no particular circumstances from which it can be inferred, survivorship is presumed from the probabilities resulting from the strength, age and sex, according to the following rules: . . . Fifth. If one be under fifteen or over sixty, land the other between those ages, the latter is presumed to have survived.”

The provisions of the statute are applicable only when it is not shown “who died first” and there are “no particular circumstances from which it can be inferred.” This is but another way of saying that the statutory presumption of survivorship is applicable only where the relative times when the commorientes died cannot be shown by direct evidence or by circumstantial evidence, or by both. (Estate of Loucks, 160 Cal. 551, 554 [Ann. Cas. 1913A, 868, 117 Pac. 673 ].) In the present case there is no direct evidence that one survived the other. But respondents claim that the evidence discloses circumstances from which it may be inferred that Effie McAllen perished before her mother.

When the statute speaks of “particular circumstances from which it can be inferred” that one died before the other, it means that there are "circumstances from which the fact of death by one before the other may be inferred as a rational conclusion from the facts proven. The statute does not mean circumstances which would show, or which would tend to show probably that one died before the other. (Grand Lodge A. O. V. W. v. Miller, 8 Cal. App. 28 [ 96 Pac. 22 ].) When, by circumstantial evidence alone, a party seeks to prove a survivorship contrary to the statutory presumption, the circumstances by which it is sought to prove the survivorship must be such as are competent and sufficient when tested by the general rules of evidence in civil eases. The inference of survivorship cannot rest upon mere surmise, speculation, or conjecture. As was said in Grand Lodge A. O. U. W. v. Miller, supra, “if the matter is left to probability, then the statute settles the presumption. ’ ’

An inference cannot be said to be established by circumstantial evidence, either in a civil or a criminal case, unless the circumstances relied upon are of such a nature and so related to each other that it is the only inference which can fairly or reasonably be drawn from them. If other inferences may reasonably be drawn from the facts in evidence, the evidence does not support the inference sought to be deduced from it. This rule is stated in Neal v. Chicago etc. Co., 129 Iowa, 5 [2 L. R. A. (N. S.) 905, 105 N. W. 197 ], as follows: “If it appears that the facts and circumstances from which a conclusion is sought to be deduced, although consistent with that theory, are equally consistent with some other theory, they do not support the theory contended for.” In Puckhaber v. Southern Pac. ,Co., 132 Cal. 366 [ 64 Pac. 480 ], the court quotes from Stringert v. Boss Tp. (Pa.), 36 Atl. 345 , as follows: “ . . . the question then arises whether the inference that the death was caused by the negligence of the defendant is the only inference that can be drawn from the facts that are in evidence. If it is not, and other inferences may be drawn from the same facts, then there is nothing for the jury but mere guesses or conjectures, and upon these no verdict can be founded.” (See, also, 10 R. C. L., p. 1007.)

With these principles in mind we proceed to a consideration of the evidence. The case is not one presenting a conflict of evidence. The evidence is without substantial conflict. And, as we have stated, there is no direct evidence that either decedent survived the other, the evidence being wholly circumstantial. The proved facts are substantially these: Ellen M. Wallace and her daughter E’ffie W. McAllen, with three other persons—a Mr. Williams and a Mr. and Mrs. Clark—were riding in an automobile which was struck by a fast traveling train of the Santa Fe Railway Company at a place where the highway is crossed by the railroad tracks. All of the five occupants of the automobile were killed. The train, which was proceeding westerly toward the city of Los Angeles, was running at a speed of fifty miles an hour. The automobile, traveling along the highway in a northerly direction, was struck by the train while crossing the railroad tracks. It was struck at about its center, or a little back of its center. An instant before the collision the fireman on the engine saw a woman who was seated on the left-hand rear seat of the automobile rise up. She was on the side of the automobile next to the engine, and was directly in the path of the onrushing train. This woman was Mrs. Effie W. McAllen,^ the identity being established by means of a photograph shown to the witness. After striking the automobile the train traveled twice its length, or approximately six hundred feet, before it was brought to a stop. The five occupants of the automobile were knocked out of the vehicle and their bodies scattered along the ground by the side of the railroad tracks. All five were dead when the first person to reach the scene of the collision arrived at the spot. The body of Mr. Will iams was found nearest to but west of the crossing. Farther west—that being the direction in which the train was traveling—Mr. Clark’s body was found. The third body from the .crossing was that of Mrs. Wallace, ‘and the fourth was that of Mrs. Clark. The remains of Mrs. McAllen were scattered along the track for quite a distance. Portions of her remains were found nearer the highway crossing than the spot where Mrs. Wallace’s body was found; other portions were found scattered along the track beyond the bodies of the four other victims. Mrs. McAllen’s body was literally dismembered, and had to be gathered up in pieces. The head was so crushed that the features were unrecognizable, and the body was ground into pieces of about twenty-five pounds weight. Her skull and face were crushed and the brains were pressed out of the top of the skull. The spinal cord was severed. The remains were identified by the clothing and rings which she had worn. The body of Mrs. Wallace was not dismembered but was cut and bruised. The neck was broken, the chest was crushed and the face was crushed in at the nose. Most of the wreckage of the automobile was carried on the cowcatcher of the engine until the train was stopped and the wreckage removed.

A licensed and practicing physician, duly qualified to give an expert opinion in such matters, testified as follows: “From my study and experience in the practice of medicine and surgery, it is my professional opinion that death would have been instantaneous in the case of a person whose body was struck by a railroad train and mangled into bits, with the spine severed and the head so badly crushed that the features were not recognizable. It is also .my professional opinion that a person receiving a broken neck, crushed chest and with head and face cut and the nose badly crushed in, occasioned from being struck by a train, would probably have also died instantly, although some cases of record show that a person has lived for some considerable length of time with a broken neck. In the great majority of eases, however, it is to be expected death ensure instantaneously from a broken neck, although some eases to the contrary are reported.”

We are unable to find in the circumstances attending the tragic death of the unfortunate victims of this shocking disaster a rational basis for the inference that one of them expired before the others. Unaided by any legal presumption of survivorship, it cannot be said that one gave up the ghost before the breath of life left the bodies of the others. It is argued by respondents that because Mrs. McAllen was seen seated on the left side of the automobile, the side nearest the engine, she must have been struck by the engine before her mother was; that therefore her spinal cord must have been severed before her mother’s neck was broken, and that since a severed spinal cord produces instant death in all cases, Mrs. McAllen must have died before her mother. We think this conclusion is non sequitwr. Unless there is reasonable ground for the inference that Mrs. McAllen’s spinal cord was severed before her mother’s neck was broken it cannot be inferred that the mother survived her daughter, even for an infinitesimal fraction of time, merely because a severed spinal cord always produces instant death while there are reported cases where death from .a broken neck did not immediately ensue. All the probabilities favor the theory that Mrs. Wallace died the instant that her neck was broken. That is the hypothesis which accords with the general rule in such eases, the medical witness having testified that “in the great majority of cases it is to be expected death ensues instantaneously from a broken neck, although some cases to the contrary are reported.” Respondents’ theory, therefore, rests entirely upon the hypothesis that Mrs. McAllen was the first to be struck, that the first impact of the engine with her body was made by the forward part of the engine and that it was then that her spinal cord was severed. But unless this hypothesis be the only one which fairly and reasonably accords with the known circumstances of the case, as shown by the evidence, it cannot be said to be established as an inference from the proved facts. For, as we have shown, it is not sufficient that the circumstances of the ease be consistent with respondents’ theory. They must be inconsistent with any other reasonable theory equally deducible therefrom.

The fact that Mrs. McAllen was seen seated on the left side of the automobile an instant before it was crashed into by the onrushing train and its five occupants hurled into eternity is entirely consistent with the theory that Mrs. Wallace was struck and her neck broken before her daughter’s spinal cord was severed or before the latter received any instantaneously fatal injury. It is quite possible that the engine, striking the body of the automobile a fraction of a second before it actually contacted the person of any of its five occupants, may have violently projected some part of the broken framework of the small vehicle ¡against Mrs. Wallace’s neck, breaking it instantly and even before Mrs. Wallace was actually hit by any part'«of the engine or by any part of the wrecked automobile. If it be objected that this hypothesis involves a too great refinement of the successive intervals of time between these supposedly succeeding events, our reply is that respondents’ position is equally chargeable with the same criticism. The fact is that the death-dealing impacts occurring between some part or parts of the engine or of the broken automobile -and the persons of these unfortunates were so nearly synchronous, in an absolute sense, that for all practical purposes it is impossible to regard them as discrete events separated in time. But if they could be visualized as separate happenings-—if those' blows which caused instant death may be regarded as occurring successively—then we say that there is no way by which it can be judicially determined that one preceded the other. For example, it is possible that Mrs. McAllen, who was seen by the fireman to rise up in the back seat of the automobile, may instinctively have hurled herself -toward the right side of the ¡automobile, -away from the onrushing engine and its train of cars, and that as a consequence the mother, who was over sixty and therefore presumably less active, was struck first and her neck broken by the engine a fraction of ¡a second before the daughter received the injury which resulted in the severing of her spinal cord, or before the latter received any injury that produced instant death. Moreover, the hypothesis advanced by respondents assumes that Mrs. McAllen’s spinal cord was severed by being struck by the .extreme front part of the engine. It is possible that she may have been hurled from the automobile while life was in her body and after her mother’s, neck was broken, and that after being so hurled from the vehicle her spinal cord was severed under the grinding wheels of the engine or by whatever part .of the train it was which mangled her body ¡and ground it into pieces. Furthermore, the evidence does not show where Mrs. Wallace was seated in the automobile. It is possible that she also was seated on the left-hand side of the automobile, i. e., on the side next to the engine. She may have been seated on the left-hand side of the front seat, directly in front of her daughter. If that was the seat occupied by the mother, then if the engine hit the automobile at a certain angle it would be possible for the mother to be struck a fraction of a second before the engine touched the daughter, notwithstanding the fact that the automobile was hit a little to the rear of its center. It follows, therefore, that the angle at which the engine hit the automobile and its occupants may be one of the essential factors which would have to be considered in any judicial determination of the question of survivorship. But the angle at which the engine struck the automobile is a matter upon which no light is thrown by the evidence. No witness gave any testimony showing what was the angle at which the engine struck the automobile. It may have struck the lighter vehicle at a right angle or at an acute angle, for aught we know.

Regardless of how the mother and daughter were seated in the automobile there would at most be but three or four feet between them; while the train, traveling at a speed of fifty miles per hour, was moving at the rate of more than seventy-three feet a second. Rushing upon the automobile and its luckless occupants at this terrific speed, the huge engine, with the combined weight of the cars behind it, snuffed out the lives of mother and daughter. In an instant, in the twinkling of an eye, all was over. Under such circumstances it were useless to speculate upon the problem of survivorship. In a second the scene passed beyond the vision of human penetration. What was said in Newell v. Nichols, 75 N. Y. 89 [ 31 Am. Rep. 424 ],, is pertinent here: “It is as unbecoming as it is idle for judicial tribunals to speculate or guess whether . . . one or the other may not have ceased to gasp first.”

It doubtless is possible to imagine a number of hypotheses, all equally consistent with the proved facts of the case and all consistent with the theory that Mrs. Wallace’s neck was broken before her daughter’s spinal cord was severed or before the latter received any injury that would cause instant death. The three or four which we have presented as possible hypotheses alone suffice to show how every theory, including respondents’, must in the end rest upon no firmer foundation than mere guesswork and filmy conjecture. Those of a speculative bent may guess or imagine or fancy, but the law requires evidence. Here there is an absolute dearth of evidence upon which to give a judicial decision that either survived the other.

It follows that the trial court, on the evidence which was presented to it, should have found the fact of survivorship according to the statutory presumption declared in subdivision 40 of section 1963 of the Code of Civil Procedure. Had it so found, appellant, as the sole heir of Mrs. Wallace’s legatee, Effie W. McAllen, would have been entitled to have the balance of his deceased mother-in-law’s estate distributed to him.

The statutory presumption of survivorship may be an arbitrary one, but the orderly disposition of property sometimes necessitates the application of arbitrary rules. In the absence of evidence from which the sequence of the deaths may be inferred as a fact, property rights necessarily must be settled .upon some theory, respecting survivorship, whether it be that of the common law, which, in the absence of evidence that one died first, disposes of the property as though the deaths had occurred simultaneously, or whether it be that of the Roman law—the prototype of our code provisions upon the subject. The legislature of this state has spoken. It has declared what presumptions shall be indulged when there is an absence of evidence from which survivorship may reasonably be inferred as a fact. Its fiat is final.

The decree is reversed.

Works, J., 'and Craig, J., concurred.

A petition by respondents to have the cause heard in the supreme court, 'after judgment in the district court of appeal, was denied by the supreme court on December 10, 1923, and the following opinion then rendered thereon:

In denying a petition for transfer to this court we place our denial of the petition for transfer upon the ground stated by the district court of appeal, namely, that the question as to whether the mother or daughter died first is not 'ascertainable from the evidence and that the finding of the trial court that the daughter died first is not sustained hy the evidence. We do not wish to be understood as approving the statement of the rule by the district court of appeal as to the method of weighing circumstantial evidence when applied to a reviewing court on appeal. The rule is stated and applied in the -recent cases of Estate of Loucks, 160 Cal. 551, 554, 555 [Ann. Cas. 1913A, 868, 117 Pac. 673 ], People v. Tom Woo, 181 Cal. 315 [ 184 Pac. 389 ], and Mah See v. North American Acc. Ins. Co., 190 Cal. 421 [26 A. L. R. 123, 213 Pac. 42 ],