Plaintiffs (a firm) owned and ran a grist mill hard by defendant’s railroad at a way station, Hobferg, in Lawrence county. In the shank of the evening of January 10, 1908, mill and content's (barring a salvage of the boiler and engine, put at $1000 to $1500) were burned. Alleging the fire caught from defendant’s locomotive, plaintiffs sued in the Lawrence Circuit Court, putting their damages at $17,163.10.
On issues joined, the jury found for defendant. From a judgment following, plaintiffs appeal — assign*68ing error in that the court permitted counsel in his opening statement to make prejudicial remarks, admitted incompetent testimony, permitted repeating prejudicial questions, erred in refusing to strike out testimony and in giving an instruction. For defendant it is argued, contra, that there is no such error. Further, in effect, that plaintiffs made no case; hence (counsel say) an asked demurrer to the evidence should have been given, and that such alleged error, if any exists, does not affect the merits or concern an appellate court.
Such, in outline, are the issues below and here.
I. The administration of justice cannot be too often or too much quickened by recourse to salutary statutes intended to produce just practical results— this, as over against the vulgar and pernicious fallacy that a law suit on appeal is a mere game of wits to be played according to highly artificial rules, over which “game” we sit as a mere umpire, according points to one player or another by the dry and lifeless rules of the game for the s.ake of the game itself. Elevated and uniform justice could not be administered without rules. If there were no rules, we would be governed by men, not laws, by discretion, a crooked metewand, not by fixed rules known to all. Order is not only Heaven’s first law, but order is of the essence of the science of jurisprudence. But rules are not the ultimate end, the main thing — that main'thing is justice itself, the very right of the matter. The rules are only in aid of that main thing — the working tools whereby it is attained.
Section 1850, Revised Statutes 1909, reads: ‘ ‘ The court shall, in every stage of the action, disregard any error or defect in the pleadings or proceedings which shall not affect the substantial rights of the adverse party; and no judgments shall be reversed or affected by reason of such error or defect.”
*69Section 2082, Revised Statutes 1909, reads: “The Supreme Court, or courts of appeal, stall not not reverse the judgment of any court, unless it shall believe that error was committed by such court against the appellant or plaintiff in error, and materially affecting the merits of the action.”
In establishing a working theory to administer those statutes, it is stiffly held that if a plaintiff is allowed all his competent proof and makes no case on the facts on which liability can be predicated and is unsuccessful below, then error in his adversary’s instructions, or in the admission of evidence on the part of such adversary, cannot “materially affect the merits of the action,” or “the substantial rights of the adverse party.” In such case (as to an appealing plaintiff), at bottom there are no merits and no substantial. rights in the eye of the law. [Trainer v. Sphalerite Mining Co., 243 Mo. 359.]
At the threshold, then, lies the question: Did plaintiffs make a case for the jury? If that question be answered, no, then mere error in instructions and in the admission of evidence on behalf of defendant is afield and fills no office at all on appeal under the quoted statutes; for the result, despite the error, was right. If right, it abides. We will not undo what has been done, in order that some intermediate move in the (so called) “game” may be corrected only to get the same result again. If, yes, then error in the way the case was put to the jury at once takes on new color and becomes of substance.
II. Of the facts. Attending to them, defendant’s right of way at Hoberg is one hundred feet in width and its track lies in the center. There is a bit of obscurity on the point but from what follows we take it the track runs southeast and northwest. Plaintiffs own a tract abutting on said right of way. Its description with other substantial testimony indicates *70that defendant’s road ran to the northwest. Plaintiff’s mill stood broadside parallel to the right of way, fonr or five feet therefrom, and,, say, fifty-four feet from the track, and must have had the same relation to the cardinal points of the compass. This has some significance, taken with the way the wind sat the night of the fire. In the afternoon and evening of January 10, 1908, there was a high wind blowing from the northwest. Obviously the same direction of track and wind is a factor in determining the course locomotive sparks would take. It is more convenient, however, to refer to the mill and track as running east and west and witnesses now and then drop into that form of expression. So, it was very cold the afternoon and evening of the fire. All agree it was foul weather-damp, misty and threatening. Some say rain fell in that region in the afternoon. Some, that it snowed in the evening, at least, during the fire. The mill, a wooden structure of sixty-barrel capacity, was newly built, partly of old material. There was a main building of three stories, with additions of an engine room and a wareroom, each one story. The wareroom ran along the entire east end of the main mill. The engine room ran along its west end, hut lacked twelve feet' of being as near the track. The main mill was covered by a composition roof — the engine and ware-rooms, with galvanized iron. The mill was completed, except wheat bins and the wareroom, and carpenters were working on them. Windows were in and doors hung except one — a large door on the north side of the main mill, fronting the track and intended to shove to and fro on rollers. That door was fastened shut on occasion with eight-penny nails. At such times it was not nailed “solid,” but was secure. On warm days they took this door away. For a week before January 10 there had been no fire in the engine room, but the mill, theretofore grinding, was open for business, occupied by the proprietors plying their *71calling as millers, and by carpenters on that day. Heat was furnished at the time by a coal oil stove in a room called the “office,” when the engine was not running. On Jannary 10 that coal oil stove was running. Witnesses vary on the time the mill men quit work and left, closing it for the evening, but it was nigh five o’clock; its windows were fastened and doors locked, except the north door and that was nailed up as usual. As we grasp it, there was a rule made by- Groh against smoking in the mill, but, as he made the rule, he accommodatingly waived it in his own favor, smoking cigars and pipes in the office. This office was in the southeast corner of the main mill, next the wareroom. Groh testified that he turned down the oil stove before leaving the mill; that his habit was to see it out before he went away. Stored in the mill were some hundred bushels of wheat, 10,000 pounds of flour, 3000 pounds of feed, 190 bushels of corn, two barrels of oil and a barrel of oil waste, the latter in the northwest corner of the mill proper, and some clothes close by. All agree that, a young man named Owens discovered the fire and first gave the alarm. He puts himself at church about two miles from Hoberg and, according to his watch, left church at about a quarter to nine. He drove in a buggy the two miles to Hoberg and discovered the fire at about a quarter after nine, when some distance away. He then drove to the Creymeyer house where Groh was sleeping (a house about one-fourth of a mile from the mill), awakened him and others and then drove north of the mill. At first he saw the blaze coming around the southwest corner of the engine room. Afterwards, when on the north of the mill, he saw the fire in the northwest corner of the main building, both in the inside and on the outside, and the weather-boarding there was burnt through. When he saw the fire at the southwest corner of the engine room it was a blaze about the size of a small brush pile. There was other testimony that Owens *72gave the alarm at Creymeyer’s house at about a quarter to nine, instead of a quarter after. At any rate, it was so late that the Creymeyer household, including G-roh, were in bed at that time. After Grroh and the others there dressed and got to the mill, as said one-fourth of a mile away, they tried to get in at the office window on the south side, but it was too smoky. They next went to the engine room and tried to get in, but it was too smoky there. Afterwards they went round to the north and west and found the fixe along the north side of the mill, more of it to the west than to the east of the main door on the north, and as near as they could make out burning most fiercely in the northwest corner. The building collapsed within a half hour after they got there. Before that there was an explosion and it seems there was a stick of giant powder with a fuse and fulminating cap in the mill, left over from digging a well. . All agree that the nailed-up door on the north was then open. It was pushed back on the top, resting against a bench in a slanting position, but was not on fire. Along the west end of the main mill, up aways, about where the roof of the engine room was to be joined to the west wall, there was a space left in the weather-boarding* when the main mill was built. Presently, when the roof of the engine room was joined to this main wall, this space was not entirely closed, and, as we understand it, there was a strip of weather-boarding off there, about three inches in width, and things were in this fix on January 10th. The railroad track ran by the mill in a little cut and this missing strip of weatherboarding would be a trifle higher than the top of a locomotive smokestack. Locomotives had been daily passing and repassing while the mill was building and shavings and other combustible debris were scattered around, but before January 10th no fire.had caught from sparks. There was testimony that fire had caught from such engines in the grass and in a rail *73pile in the neighborhood at uncertain times prior to this fire, some at a less and some at a greater distance away than the mill stood from the track; bnt no one had ever seen an ember smoulder for three hours. All of such fires were observable within a short time after the passage of a locomotive. There was also testimony that at some uncertain time after the fire in the mill sparks now and then fell from passing locomotives as far as the mill basement. Plaintiff’s theory is that the fire started about six p. m. from sparks thrown by a locomotive heading west in the teeth of the wind and drawing a light passenger train of three or four cars. There is no testimony that this particular engine set any of the other mentioned fires. To the contrary, it had been in the shop, was overhauled there, and came into use in prime condition about two weeks before the fire. It is established by the proof, with no countervailing testimony, that the state of the weather (whether cold, hot, dry, damp or heavy) has a great deal to do with the life of sparks thrown from a locomotive; that when it is very cold or damp sparks do not live as long or fly as far as when it is warm or dry. Weather conditions existing at the times the other -fires started or sparks fell, testified to by witnesses, are not disclosed. There is testimony that all locomotive engines throw some sparks. The plan in use on defendant’s locomotives to prevent the escape of coals of fire, as distinguished from sparks, is by means of a spark arrester. Such arrester is a composite device having a baffling sheet and a steel sheet or sieve with very small meshes, three thirty-seconds, of an inch in diameter. This appliance on this engine was in good order at the time. When the draft is on the fire box so there is a tendency to lift coals of fire, such coals first strike the baffling sheet and are there mechanically pounded and pulverized by attrition and beating into a less size than the small meshes in the sieve before they can escape by *74passing through it and out of the smokestack into the air. The station at Hoherg is six or seven hundred feet east of the mill. The train in question stopped at the station January 10th at about five forty-eight p. m., the usual time. There was some testimony it was on time. But plaintiffs’ testimony indicates it was a little late, say, five or ten minutes, making it about six o’clock when it passed the mill. If the train was on time, then on Owens’ testimony the fire was discovered over three hours after its passage. If a little late, as plaintiffs’ evidence shows (and we must assume it was on demurrer), then it was nearly three hours before the fire was discovered. There is no testimony contradicting Owens on the size of the fire at the time he first saw it, and there are no facts shown making it improbable, in the keen gale then blowing, that thereafter a fire, of the size seen by Owens, could not have reached the fury and volume shown by the other witnesses when (getting' out of bed, dressing and going on foot one-fourth of a mile) they got to the mill. "When working steam up grade there are more sparks thrown by a locomotive then when running down grade, and more sparks from an old fire in the fire box than from a green fire. It was usual in starting the train west from the station house at Hoherg to make a green fire by shoveling in fresh coal and the steam is worked to gather headway after the start. At the mill there would be a light pressure of steam. There is not a particle of testimony that the particular engine drawing this passenger train, leaving the Hoherg station at, or a little before, six p. m., on January 10th, threw any sparks. No witness saw it throw any, though they saw the train and engine, and the testimony is to the effect that in the nature of things it could throw off very few, if any, and those not amounting to much. Out of forty or fifty times one witness had seen it go by, he saw it throw sparks four or five times, hut whether before or after the *75fire is dark. Mr. Groli said at tie fire tliat somebody bad broken that door open (referring to the nailed np door on the north side). A witness said to him: “It has been'too long for a train to have set it.” To that fie replied: “Yes, there has been no train for over three hours.” He told another witness he did not know how the fire originated. Plaintiffs’ attorneys elicited from one of their own witnesses that there was talk at the fire to the effect it was of incendiary origin and that there had been trouble with reference to digging a well at the mill. A witness for plaintiffs, who went to the post office after supper to get mail brought in on the train, had a chance to see and saw no fire at the mill. Another, who lived not far from the mill and whose bedroom window faced the north side of the mill, saw no fire when he retired about an hour and three-quarters after the train had passed.-'
Defendant was allowed to show, over objection, that a stranger came to defendant’s section house from the direction of the mill that evening after the train passed and wanted work, said he would have to lie out, borrowed some matches to build a fire and returned in the direction of the mill. Defendant, over objection, at first was allowed to show that the mill was insured, that the insurance was paid after the fire, that the policy had a subrogation clause in favor of the company'if a fire was caused by á railroad, and that the insurance company was interesting itself in the prosecution of the suit. Later the court changed position and the same character of testimony was ruled out. • Plaintiffs got an instruction commanding the jury to disregard the matter of insurance. Defendant was permitted to show by experts, over objection, how far a spark from a locomotive under prevailing weather conditions could fly and live, and how far a spark would have to go to reach the mill, keeping in view the direction of the wind and the distance of the track, The details of all the aforesaid testimony *76may be omitted'. Nor we need not say on demurrer whether there was error in that behalf, since, as already held, if plaintiffs fail to make a case, such rulings could not be reversible error.
On-such record, we rule as follows:
(a) The action is brought under the statute (R. S. 1909, sec. 3151) enacted in 1887 (Laws 1887, p. 101). G-rounded in natural equity and held a constitutional exercise of the police power (Mathews v. Railroad, 121 Mo. l. c. 315), resting also on the maxim, “Every one should so use his own property as not to injure that of his neighbor” (Campbell v. Railroad, 121 Mo. l. c. 346), that statute with a single stroke cut away all necessity of pleading or proving negligence in order to fasten liability on a railway company for damages from fire escaping from its engines. It was a revolution in existing law, was remedial, and is to be liberally construed. It runs on the theory that such company, speeding its trains through the country " by steam power, in using so good a servant as fire when kept in bounds, but so hard a master when set free, was henceforth responsible for the ravages of a fire set out by its locomotives. It means that, regardless of fault, the one must pay the damages who caused loss by that particular kind of a fire. To prove the negligent escape of fire from a locomotive, here one instant and gone the next, and manned by unknown men, was an almost impossible task, and the law-maker took off that burden. So far as material, the statute reads: “Each railroad corporation owning or operating a railroad in this State shall be responsible in damages to every person and corporation whose property may be injured or destroyed by fire communicated directly or indirectly by locomotive engines in use upon the railroad owned or operated by such railroad corporation. . . .”
Observe, while negligence is no longer a factor, and plaintiff is relieved from the burden of charg*77ing or proving it, yet lie still carries a burden. He cannot recover by mere proof of a fire and tbe presence of locomotives on a railroad. He must still charge and prove that the fire was actually communicated to his property by one of defendant’s locomotives in use on its road. Failing in that, he is cast as a matter of law.
(b) On demurrer to the evidence, the right doctrine to go by is: Defendant’s testimony, where contradicted, is taken as false; plaintiff’s, whether contradicted or not, is taken as true; discrepancies, contradictions between witnesses or self contradictions by witness, together with the credibility of the witness and the weight due his testimony, are for the jury, not the court. So, plaintiff is entitled to the grace of having allowed in his favor every inference springing reasonably on the proof.
Not unmindful of those precepts, but applying them in full vigor, we are of opinion the facts disclosed were not sufficient to carry the case to the jury.
This, because:
(c) The facts have been set forth so fully that there is no call to restate them and little to comment on them.
(1) The evidence relied on to show the origin of the fire is strictly circumstantial — that is, the main fact, viz., the cause of the fire, stands to be proved by the proof of surrounding conditions and circumstances “whose existence.is a premise from which the existence of the principal fact may be concluded by the necessary laws of reasoning.” (Bl. L. Dict., under “Circumstantial Evidence.”) “The soundest test of the validity of that sort of evidence is, that no other theory but the hypothesis upon which the conclusion is based, can be formed.” [Per Hughes, J., in Musselwhite v. Receivers, 4 Hughes (U. S. Cir. Ct. Rep.) l. c. 169.] In circumstantial evidence the principal and ultimate fact is got at by way of argument and *78by method of demonstration in. the nature of reductio ad absurdum. [Will’s Cir. Ev., p. 17.] Therefore, in cases turning on circumstantial (or what is called by the civilians oblique and by the Scotch jurists argumentative) evidence, the proof should have a tendency to exclude any other reasonable conclusion than the principal fact. Says the author just quoted (p. 17): “The force .and effect of circumstantial evidence depend upon its incompatibility with, and incapability of, explanation or solution upon any other supposition than that of the truth of the fact which it is adduced to prove.”
(2) Testing the facts by the general principles announced we see no way out of it except to hold that plaintiffs did not show that defendant’s locomotive directly or indirectly communicated fire to their mill. There is not a particle of evidence tending to show that a spark, keeping in its whole course so alive and hot as to breed a fire when it lit, could fly or ever flew as far as such spark would have to carry in this instance and in a gale of as cold and damp a wind. There is no testimony tending to show that a fire would lurk and smoulder for the length of time this must have done, if it is to be credited to defendant’s locomotive. Common observation does not teach us that fires, catching in tinder or debris, hesitate so long. Usually in cases finding their way into courts for damages for setting out fires, they develop shortly after a locomotive has passed, and the significance of time has been a topic of judicial comment in those cases. The whole trend and thread of the proof point to the conclusion" that the fire originated inside the mill. Indeed, in damp conditions prevailing, it could not in reason have caught on the outside, therefore the spark must have got inside the mill. To account for a spark falling inside, a three-inch crevice in the weather-boarding on the west side is shown. Where it led to in the inside is not shown. Did it lead to the floor. *79to a sill or to the ground? Assuming it led into the main mill, then according to the physics of the matter, a spark (from a locomotive running to the northwest in the teeth of the gale) would have to travel nearly on a dead level for a great many more than fifty-four feet before it could reach that particular hole in the west wall, for we know the smokestack would have to he smartly to the west of the mill, before a wind from the northwest could carry the spark to that hole. That such spark would live in the cold and damp, reach that crevice and breed so late a fire inside of the mill rests necessarily in speculation, and has all the loose and inconclusive features earmarking guesswork and conjecture as distinguished from sound reasoning. Did tramps inadvertently or some enemy intentionally set the fire? Did mice gnaw matches in some coat pocket, or did the coal-oil stove explode because turned down low and not out? Or did the barrel of waste set fire to the mill by spontaneous combustion? Who knows ? Strange things might happen in that way, and whether they did or not would be conjecture pure and simple. But some of those strange things would be just as reasonable as to say that a spark from a locomotive (that was not seen to throw any) flew on a level for, say, seventy-five or one hundred feet and accurately hit a three-inch crevice in a wall; that it then was red hot, though the air was cold and damp; that it then fell on combustibles inside the mill and started a fire; and that such fire smouldered for three hours (a little more or less) before discovery by those situated so as to likely see it. That hypothesis is also conjecture pure and simple. No conclusion that such result was likely or reasonable could flow from sound reasoning of such premises. At very best one could only say, it was possible, it might be so. Shall we say the proof shows that defendant might have set the fire or might not, ergo, defendant must pay? On such theory, defendant would always have to pay, and the *80statutory duty of showing that defendant’s locomotive communicated the fire would he eliminated. We could not very well adjudicate on such a toss up and non sequitur as that. Speaking of cold weather as discouraging to long life in a spark of fire, who has not got up of a bitter morning to find his household fire frozen out? At such times one is — but no matter about that.
It is true that testimony is permitted in this character of eases to show the ability of locomotives to throw sparks to this, that or the other distance, and, when so thrown, to show they ignite combustible matter. But the value of that sort of testimony, in the nature of things, should be limited to similar conditions of wind and weather and surrounding circumstances. We know nothing of the state of the weather, or the direction of the wind, or the character of the combustible material (except grass) set on fire theretofore in that region by locomotive sparks. The proof fell short on that score. So, there is no testimony whatever that the engine in question threw any sparks on the evening of the 10th of January, 1908, likely to cause a fire. So, does not the fact that the door on the north was found pushed in, point to the probable presence of intruders in the mill after closing time?*
If the jury had found for plaintiffs, it would have been the duty of the trial court to set the verdict aside as unsupported by proof. Under facts more persuasive than here, that result has been reached more than once. [Gibbs v. Railroad, 104 Mo. App. 276; Manning *81v. Railroad, 137 Mo. App. 631; Peck v. Railroad, 31 Mo. App. 123; Bank v. Railroad, 98 Mo. App. 330; Funk v. Railroad, 123 Mo. App. 169; Otis Co. v. Railroad, 112 Mo. 630-1.]
In the foregoing view, other questions are not reached and will he reserved.
The judgment was right. Let it be affirmed. It is so ordered.
All concur; Graves, P. J., in separate opinion.