(after stating the facts.) Several questions are raised in the record as. to alleged errors in not permitting certain witnesses to answer certain questions propounded to them by the appellant, the answers to which, it is contended, would have had a tendency to discredit or impeach such witnesses before the jury, and it is also contended by appellant that the .trial court erred in giving certain instructions asked by appellee; but, in view of the opinion of the majority of the court, it is unnecessary to pass upon these questions. The real question to be considered and determined is .as to whether the evidence is legally sufficient to sustain the verdict of the jury.
Under the facts in this case, it was the duty of the oil company to use ordinary care in transferring or emptying the gasoline delivered by it into the receiving box on Exchange Street; and if it failed to exercise such care, and the plaintiff was injured thereby, the oil company would be liable, if such injury was the direct result of such want of care, or if such injuries could have been reasonably anticipated or foreseen as the probable result of such want of care of negligent act; otherwise it would not be liable.
In Derry v. Flitner, 118 Mass. 131, the court said: “The rule is well settled, and is constantly applied in this commonwealth, that one who commits a tortious act is liable for an injury which is the natural and probable consequence of his misconduct. He is liable not only for those injuries that are caused directly and immediately by his acts, but also for such consequential injuries as, according to the common experience of men, are likely to result from his act. * * * The true inquiry is, whether the injury sustained was such 'as, according to common experience and the usual course of events, might be reasonably anticipated.” In Braun v. Craven, 51 N. E. 657, the court said: *617’“'That before the plaintiff can recover he must show a damage naturally and reasonably arising from the negligent act, and reasonably to be anticipated as the result.” In Hoadley v. Northern Transportation Co., 115 Mass. 304, the court said: ' “The legal damages that follow any wrong are only such as, according to common experience and the usual course of events, might reasonably be anticipated. The defendant’s liability extends only to natural and probable consequences.” In Hoag v. Railroad Co., 85 Pa. St. 293, the court said: “In determining what is proximate cause, the true rule is that the injury must be the natural and probable consequence of the negligence; such a consequence as, under the surrounding circumstances of the case, might and ought to have been foreseen by the wrongdoer as likely to flow from his acts.” In Chicago, St. P., M. & O. Ry. Co. v. Elliott, 5 C. C. A. 349, the court said: “An injury that is the natural and probable consequence of an act of negligence is actionable, but an injury that could not have been foreseen or reasonably anticipated as the probable result of the negligence is not actionable.”
The burden of proving that the oil company was negligent in transferring or emptying the gasoline into the receiving box was upon the plaintiff. Negligence is never presumed, but like any other fact must be proved. Hence it necessarily follows that to entitle the plaintiff to recover he must show that the oil company was negligent in the matter of the delivery of the gasoline into the receiving box, and that his injuries were the direct result of such negligent act, or that such injuries might in the usual course of things have been reasonably anticipated or foreseen.
It is contended by the appellee (plaintiff in the court below) that the oil company was negligent in emptying the gasoline into the receiving box, that in the use of a paper funnel it became crumpled and stopped the flow of the gasoline to some extent through the receiving pipe, and, therefore, a part of the gasoline flowed into the receiving box until it was filled to a point above the other small nipple or pipe in the receiving box, and then made its way down this little pipe to the earth, then following the track or line of pipe connecting the receiving box with the storage tank through the soil some 10 or 12 inches to the stone retaining wall, then through the retaining wall into the. coal house, then it *618flowed down the wall a foot or two to the left or old door with one end inserted in the retaining wall, the other end extending obliquely downward to or near the inner wall of the east side of the coal house, then to the bottom of the said inner wall, then through the double wall of the coal house, then on to the granitoid floor, then flowing between the south end of the mound covering the tank and the south wall of the open area, then under the engine house, coming out at the north end, continuing down to the grated sink in the northeast corner of the area, and in a few moments after the oil was seen thus flowing the explosion occurred, wrecking the pool room where the plaintiff was at the time with perhaps 150 or 200 other persons, and the plaintiff was injured as alleged.
The oil company controverts this contention. It denies that it was negligent in any way whatever in emptying the gasoline into the receiving box. It contends, first, that all the gasoline which was taken from the barrel was conveyed into the receiving pipe in the receiving box; second, that, even if the gasoline had flowed into the receiving box and above the opening of the other small nipple or pipe, in the very nature of things it could never have found its way into the open area as contended by appellee; third, that, even if it should have reached the area in this way, Murray, who delivered the gasoline for the oil company, could not have reasonably anticipated or foreseen such a result, and, therefore, the oil company is not liable for the injuries complained of; and, the oil company further contends that the gasoline which caused the explosion escaped from the T-pipe or from some leak in the engine or generator while in the process of being emptied into the receiving box; that gasoline is exceedingly volatile, and when subjected to great commotion, it readily forms a gas, and as it was turned into the receiving pipe to flow through the pipe connecting the receiving box with the storage tank, its tendency was to form a gas, which forced a part of the gasoline to flow or spurt out at the top of the T-pipe, and then run under the engine house and over the granitoid floor to the grated sink.
In the effort to sustain these conflicting theories, a great deal of testimony was taken on both sides, and to understand and give this testimony its ¡proper bearing upon the issues thus *619presented has entailed upon the court much time and care. It appears from the testimony that when the driver, Murray, reached the Turf Exchange premises with the second barrel of gasoline, there was 15 gallons in the storage tank, that after the explosion there was 40 gallons in it, thus showing that 25 gallons of the second delivery was in the tank. There was 53 gallons in the barrel. When it was returned there was 17^ gallons in it, thus showing that the driver Murray had drawn 35^ gallons from the barrel, 25 gallons of which was in the tank after the explosion, while 10^2 gallons found its way into the open area, and its flow was seen coming from under the north end of the engine house and extended down to the opening in the northeast corner of the area. This wás the gasoline which caused the explosion.
Now, the question is, how did the gasoline escape, and how did it reach the open area? The oil company introduced a number of witnesses showing that it had made two or more tests with tank and pipe like the one at the Turf Exchange under conditions as near as practicable the same as there, the only difference perhaps being in the temperature, and it was shown at all of these tests that when gasoline was emptied into the receiving box it would bubble or spurt out of the T-pipe at intervals at different heights from an inch to 2 feet high; that in putting in 25 gallons, 10 gallons of. it would escape through the T-pipe. One of these tests was made publicly at the courthouse in Hot Springs at which everybody was invited, and among those present were the .learned counsel for the appellee. On the other hand, the testimony shows that the gas company also made two or more tests of similar character, but the conditions were not the same as they existed at the Turf Exchange. At these tests, the bottom of the barrel was only raised high enough to siphon it out of the barrel into the receiving box. The hose used was 10 or 12 feet long. The bottom of the barrel from which the gasoline was taken was about a. foot and a half- higher than the tank, and on a level with the receiving box, thus showing that the gasoline in flowing from the barrel into the receiving box and into the tank through a hose 10 or 12 feet long only had a fall of-about one foot and one half, while at the test made by the oil company the gasoline was taken from a barrel which stood in the wagon *620some 2.y2 or 3 feet above the receiving box, and when it reached the receiving box it was conveyed through a line of pipe 19^ feet long, about 12 feet of which was horizontal and 7 feet 6 inches vertical, and adding to this the 2y2 feet, the barrel stood above the receiving box, the gasoline had a fall of 10 feet. At the test made by the gas company, it was shown that no oil would escape from the T-pipe.
While we do not intend any harsh criticism as to the manner in which the tests were made by the gas company, it is perfectly obvious that they were not made under the same conditions as existed at the Turf Exchange, and can serve but little purpose in solving the question at issue. The height of the source of supply above the tank would have some effect on the flow of the gasoline. The higher the source of supply, the greater would be the weight of the column, and the gasoline would thereby be forced more rapidly into the storage tank, causing great commotion, and thus increase the tendency to form gasoline vapor or gas, and thereby force the gasoline out of the T-pipe. The oil company made further tests by pouring first about 20 gallons of water in the space occupied by the receiving box on the outside of the retaining wall, and none of it found its way into the coal house, but the water flowed south and west. Then, waiting about ten minutes, they discharged 20 gallons of water through á siphon against the retaining wall, the west side of the coal house. Then, waiting 10 minutes longer, 20 gallons were discharged against the inside of the east wall, and 20 gallons in the southeast corner of the coal house, thus making 60 gallons of water turned into the coal house, and none of it found its way to the granitoid pavement in the open area. There was a little coal in the coal house, the most of it slack.
On the other hand, the colored porter, Arthur Harris, testified in behalf of the appellee to the effect that when the driver Murray began to empty the gasoline into the receiving pipe, he (Harris) went down to the storage tank, took the cap off of the T-pipe, and sat right over it to watch the indicator so as to notify Murray when the tank was full; that he watched the indicator; that it never moved while he was there, and no gasoline came out of the T-pipe. The evidence tends to show, however, that after the explosion the indicator stick was .in perfect order, and that in *621putting in as much as 25 gallons of gasoline, the indicator stick would rise several inches.
It is contended, however, by appellee that the evidence only shows two places at which the gasoline could have escaped; one at the rceeiving box, 'the other at the T-pipe, and the porter, Arthur Harris, having testified that it did not escapé at the T-pipe, the jury had .a right to believe him, and that their finding is conclusive, and that the effect of this finding is to establish the fact that the gasoline escaped at the receiving box, and made . its way to the granitoid floor.
If the determination of the question as to whether the evidence is legally sufficient to sustain the verdict depended solely upon the testimony of the porter, Arthur Harris, and the driver Murray, who testified that he exercised due care in transferring the gasoline from the barrel to the receiving box, that none of it escaped into the receiving box but all of it was emptied into the receiving pipe, our duty would be at an end. We should affirm the case upon the doctrine that this court will not disturb a verdict based upon the weight of evidence. But, in determining this question, we must look also to the physical facts disclosed by the record about which there is no dispute, and if, upon a full and fair consideration of these facts, to sustain the verdict would require the court to believe that which in the very nature of things could not be true, to believe that which is contrary to human experience and common observation, then it would be-the duty of the court to reverse the case, upon the ground that the evidence is not sufficient to sustain the verdict.
It has been the settled policy of this court, in'passing upon this question, to give the strongest probative force to all the testimony tending to support the verdict, and to indulge in every presumption which could be fairly drawn from such testimony, and when the evidence is conflicting, or when the facts are of such character that different minds might honestly draw different conclusions, the verdict of the jury is upheld. In other words,' a case should nót be reversed for the want of testimony to support the verdict unless the conclusion follows, as a matter of law, that no recovery can be had upon any view which could be properly or reasonably taken of the facts the evidence tends to establish, and the court has no inclination or disposition to modify or *622change this ruling. This doctrine is sanctioned and approved by the Supreme Court of the United States and by nearly or quite all of the courts of last resort in the States. It is, however, often difficult to apply a principle of law to the facts of any given case; hence the variety of expressions used by the courts such as:
“The case will not be reversed where there is a conflict in the evidence.”
“A verdict will not be set aside if the evidence is legally -sufficient to sustain it.”
“A verdict will not be % disturbed because it is against the weight of evidence.”
“A verdict supported by the evidence is conclusive.”
“A finding of the jury is conclusive, though against the weight of evidence.”
“Where the evidence is not sufficient to support a verdict, a new trial will be awarded.”
“Where there is no evidence to support the verdict, a new trial will be granted,” and many other like expressions.
These expressions do not convey a concrete, definite idea which can be applied to every given case, but they all tend alike to the same result. The thought intended to be conveyed in them is that the jury should be permitted to return a verdict according to its own view of the facts, unless, upon an honest consideration of the whole evidence and giving effect to every inference to be fairly and reasonably drawn from it, the case is manifestly for the party asking a peremptory instruction. When a case reaches this point, it then becomes a question of law for the court, and not one of fact for the jury. St. Louis, I. M. & S. Ry. Co. v. Martin, 61 Ark. 549; Hot Springs St. Rd. Co. v. Hildreth, 72 Ark. 572.
In St. Louis, I. M. & S. Ry. Co. v. Rice, 51 Ark. at p. 476, Justice Sanders, in delivering the opinion of the court, said: “It is the settled policy of this court to uphold the verdicts of juries where they have passed upon disputed matters of fact, provided the evidence be legally sufficient to support their finding. Of this it is the province of the court to judge.”
In Catlett v. Railway Company, 57 Ark. 466, Chief Justice Cockrirr, speaking for the court, said: “The Constitution pro*623vides that ‘judges shall not charge juries with regard to matters of fact, but shall declare the law.’ Art. 7, § 23. This provision shears the judge of a part of his magisterial functions, but it confers no new power upon the jury. It was the jury’s province, before this provision was ordained, to pass only upon questions of fact about which there was some real conflict in the testimony, or where more than one inference could reasonably be drawn from the evidence. The Constitution has not altered their province. It commands the judge to permit them to arrive at their conclusion without any suggestion from him as to his opinion about the facts. As Judge Battle expressed it in Sharp v. State, 51 Ark. 155, ‘the manifest object of this prohibition was to give the parties to the trial the full benefit of the judgment of the jury as to facts, unbiased and unaffected by the opinion of judges.’ If there is no evidence to sustain an issue of fact, the judge only declares the law when he tells the jury so. ‘The legal sufficiency of proof and the moral weight of legally sufficient proof are very distinct in legal idea.’ The first lies within the province of the court, the last within the province of the jury.”
Under these familiar rules, when all the evidence in this case is fairly considered and giving to it its strongest probative force and every inference which can be fairly drawn from it, we do not think the evidence is legally sufficient to support the verdict. While the evidence tends to show that the soil immediately below the small nipple or pipe in the receiving box through which it is contended the gasoline which caused the explosion passed was firm and solid, and it may be conceded that the soil -immediately under this pipe and the soil filling around the receiving pipe from the receiving -box to the. stone retaining wall was open or porous, and it may be further conceded that ten or more gallons of gasoline found its way into the small nipple or pipe, yet, in the very nature of things, the gasoline which caused the explosion could not have passed along the line of pipe through or on the soil to the retaining wall, then through and down the wall to the loft or door, then along the loft or door to the east wall of the coal house, then down the wall to the board floor, then through the double wall of the coal house to the granitoid floor in the open area. Such a contention would be not only highly improbable, but it would be irrational and at war with the physi*624cal facts, and contrary to all human experience and common information. Hence, the testimony of Arthur Harris can have no probative force, when it is in direct conflict with the conceded or physical facts, and to believe it would involve an absurdity in reason and common experience.
This doctrine has been frequently applied. In Continental Life Ins. Co. v. Yung (Ind.), 3 Am. St. Rep. 630, the court said: “That the evidence which tends to support the finding may be contradicted does not justify this court in ordering a new trial. Where competent evidence appears in the record which, if believed, necessarily tends to support the finding, unless the evidence relied on is of such character as that to believe it would necessarily involve an absurdity in reason or an impossibility according to the very nature of things, this court can not say, however much such evidence may be opposed by other testimony, that it is conclusively contradicted.”
The question has perhaps arisen more frequently in suits tor personal injuries received at railroad crossings. Suppose the-plaintiff in such a case testifies that his sense of sight and hearing are good, that just before stepping on the track he stopped, looked and listened for an approaching train, and neither saw nor heard one, that the day was clear and bright, and after stopping, looking and listening he immediately attempted to step on the track and was injured by the train, when the physical or conceded facts showed that the track at the place of the accident was straight and level for a distance of a quarter or half of a mile, and there was nothing to obstruct his view and nothing to prevent his hearing an approaching train. Could the court hesitate a moment to set aside a verdict based upon such testimony ? Certainly not. Why? Because such evidence would be conclusively contradicted by the physical facts, and would be entitled to no probative force whatever. Stafford, v. Chippewa Valley Elec. R. Co., 110 Wis. 331; Marland v. Pittsburg & L. E. R. Co., 123 Pa. 487, 490; Myers v. B. & O. R. Co., 150 Pa. 386, 390; Lake Erie & W. Rd. Co. v. Stick, 143 Ind. 449, 459, 463; Kelsay v. Missouri Pac. Ry. Co., 30 S. W. 339; Payne v. Chicago & A. R. Co., 38 S. W. 308, 311, 312, 313, 314.
In Payne v. Chicago & A. R. Co., supra, 311-2-3, the court said: “The headlight was brightly burning. The bell was ring*625ing, and the train was on time. The engineer and fireman were at their respective posts. The train was making the usual noise, and the plaintiff was expecting that very train. But he, alone, oí all others, when he went to cross the track, though a bright, active boy, with a good mind and eyesight, and acquainted with the locality and the dangers incident to crossing the track, though he listened, could not hear the whistle blown, nor the bell constantly ringing, nor the rumble of the train, nor see, with his good, young eyes, the blazing headlight of the train as it rapidly advanced upon him. Such testimony as this is so contrary to the daily experience of common life, so at war with the conceded and indisputable physical facts in this case, that neither courts nor juries can, without stultifying themselves, yield to it an iota of probative force or effect. It is a proposition too monstrously improbable for rational human belief. To argue to the contrary of this is to endeavor the transmutation of the impossible into possibility. * * * But in this instance it is plainly proved, beyond peradventure, that the statement of plaintiff ‘that he did all in his power to ascertain whether there were any trains approaching/ etc., was not, and, indeed, could not, be true. This matter of denying probative force even to direct and affirmative testimony, when such testimony is plainly at war with the physical facts and surroundings, has passed into precedent. Thus, in the leading case of Artz v. Railroad Co., 34 Iowa, 153, it is said:’ ‘But it is urged by the appellee’s counsel that the plaintiff testified that he did both look and listen to see and hear the train, but did not; and.that this testimony shows that he was not guilty of contributory negligence, or, at the very least, it made that a question of fact for the jury. The difficulty, however, with the position is that, the conceded or undisputed facts being true, this testimony can not, in the very nature of things, be true. It constitutes, therefore, no conflict. Suppose the fact is conceded that the sun was shining bright and clear at a specified, time, and a witness having good eyes should testify that, at the time, he looked, and did not see it shine. Could the testimony be true? The witness may have been told that it was necessary to prove in the case that he did look, and did not see the sun shine; he may have thought of it with a desire that it should have been so; he may have made himself first believe it was so, and *626this belief may have ripened into a conviction of its verity; and, possibly, he even may testify to 'it in the self-consciousness of integrity. But, after all, in the very nature of things, it can not be true, and hence can not, in the law, form any basis for a conflict upon which to rest a verdict.’ ”
The same doctrine has been frequently upheld in criminal cases where life or liberty was involved. Suppose a man was to be tried upon a charge of murder, and at the trial half a dozen witnesses were to swear that they witnessed the homicide through a solid brick or stone wall three feet thick, 100 feet long and 50 feet high with no openings -in it; that the defendant was the aggressor; that he brought on the attack and stabbed or shot the deceased to death. In other words, that the killing was a coldblooded murder. While, on the other hand, the defendant alone testified that the deceased was the aggressor; that he brought on the attack, and the killing was done in actual self-defense; and upon this testimony alone the defendant was convicted. Could such a verdict stand? Certainly not. Why so? Because the finding of the jury would be in. direct conflict with the physical facts. The testimony of the six witnesses would have no probative force. It would not even amount to what is called “a scintilla of evidence,” and the court could not sustain such a verdict without closing its eyes to reason and common experience, ‘and stultifying itself. This it should not do. State v. Anderson, 1 S. W. 140; State v. Turlington, 15 S. W. 141.
So it is here. If it be conceded that the only two places where the gasoline could have escaped were at the receiving box and at the T-pipe, and the porter, Arthur Harris testified that it did not escape at the T-pipe, his testimony is not entitled to any credit because he is contradicted by the physical facts about which there can be no doubt.
In State v. Turlington, supra, 147, the court said: “All the other facts and circumstances point to the same conclusion. Defendant’s words can not be believed, when contradicted, as they are, by the physical facts. Neither courts nor juries should be required to base their actions or beliefs on physical impossibilities.”
This view of the case is strengthened by the other testimony in the case. In the first place, gasoline has a tendency to rapid *627evaporation, and if it was poured into the small nipple or pipe in the receiving box it would have been rapidly absorbed and deflected through every crack and crevice over which it flowed, and the process of evaporation would have been so rapid that no part of it would have ever reached the granitoid floor. In the second place the experiments made by the oil company with pipe and tank and all equipment under substantially the same conditions as existed at the Turf Exchange on the day of the explosion show that, with the T-pipe uncovered, the gasoline would escape through it in .greater or less quantities during th'e process of emptying it into the storage tank. And, in the third place, the pouring water in the coal house at the points where it is contended the oil passed and none of it reached the granitoid floor furnishes a very strong and potent reason that the gasoline poured into the small nipple or pipe in the receiving box could never have reached the granitoid floor below.
It follows from these views that this case must be reversed and remanded, and it is so ordered.
Wood, J., disqualified.