(dissenting). — Action by plaintiff, the widow of Carl E. Puchs, deceased, to recover damages for the death of her husband, caused by the explosion of “Mill Creek Sewer.”
The petition after formal and preliminary statements alleges that the defendant city had built the sewer, and then proceeds to state that a fire broke out on the premises of defendant the Waters-Pierce Oil Company on the twenty-second of July, 1892, where a large stock of oils was stored by that company, and that-., such company did cause and permit said oils to escape and run into the above, mentioned sewer, fill the same with oil, and generate gases therein, etc.
Having made these allegations, the petition then avers that: “Under a license from the then owners of said lot (referring to the lot afterward bought by Carl E. Puchs in May, 1884) the said sewer was by said city constructed and carried underneath said lot eastwardly toward the said river, and that said sewer was located below the cellar afterward caused to be built upon said lot by Said deceased Carl E. Puchs; that when defendant, the city of St. Louis, obtained said license from said owners, it assumed and agreed with the then owners of said lot and their assigns and became bound to keep and maintain said sewer in good order and to care for the said sewer, so that- said lot and any im*183provements which might be put thereon would be free from danger of injury on account of said sewer and the use thereof.”
The petition then concludes thus: “Plaintiff further alleges that said sewer was provided with openings especially designed to carry off any gases which might arise in said sewer and be liable to combustion and explosion and that said sewer and the openings thereof aforesaid on and prior to the said twenty-sixth day of July, 1892,'were in the sole care and control of defendant, the city of St. Louis, its agents and servants, yet the said city, its agents and servants, knowing . that said defendant, the Waters-Pieree Oil Company, had flooded said sewer with oil, neglected to open said vents and carelessly and negligently failed to take measures and precautions to prevent gases arising and accumulating in said sewer so as to endanger the same; and that between the said twenty-second and twenty-sixth days of July, 1892, gases did arise and accumulate in said sewer in- great and very dangerous quantities and on the date last named, and within six months next before the commencement of this suit, ignited and exploded with great force, throwing open said sewer underneath the property of said Carl E. Fuchs, shattering his said building, and also then and there causing the death of said Carl E. Fuchs,” etc., etc.
The answer was a general denial by defendant city as well as by defendant company.
The evidence, in substance, so far as necessary to state it, was to this effect: Carl E. Fuchs, deceased, owned the building on the east side of Fourth street, seven or eight doors south of Chouteau avenue and about four blocks from the river. This section of the city is a valley and the sewer in question is known as “Mill Creek Sewer.” This sewer was formerly a creek *184and constituted the natural drainage of a large portion of the city, into which very numerous smaller sewers emptied, and it drained the property in Mill Creek valley from Grand avenue eastwardly, and also drained the city hall and Four Courts. Fuchs’ building was located over this sewer, which was built in 1858 or 1859 in the most solid and substantial manner, the stones composing it being very massive, and it ran through the lot on which the store building of Fuchs was situate, in a northwesterly and southeasterly direction, and crossed Fourth street and Broadway, which converged at that point, and were in consequence of such convergence, some two hundred or three hundred feet wide at that point.
On July 26, 1892, the sewer exploded about 4:25 p. m. and in consequence of which, Carl E. Fuchs died on that day. The explosion tore out the front of the store except the iron pillars, also the rear wall mf the entire building and the floor of the store; and opened the sewer through the whole length of the building and extended eastwardly between Second and Main streets, where the entire top arch of the sewer was thrown out for a distan ce of about four hundred or five hundred feet. The street west of .the store was not disturbed, but the ,sidewalk on the west side of Broadway was torn up and also the property next to it. There were covers for the sewer in the middle of the street (where Broadway and Fourth street join) opposite the store, and another on the west side of Broadway about one hundred and fifty feet from the store; this cover was blown off. The one in the street west of the store was an ordinary size manhole, three feet in diameter, with a solid cast-iron lid about three fourths of an inch thick; after the explosion this lid was found broken in pieces, and the contents of the store — barrels, boxes, bottles, shelving and woodwork, wood floors, joists, *185plaster, and wainscoting — were in the sewer through which the water was rushing. There was a substance in the cellar which looked like an oily mass and had a gaseous smell.
The Waters-Pierce Oil Company’s place of business was between Gratiot street on the south, Twelfth street .on the east, the railroad tracks on the north, and Fourteenth street on the west, and was ten blocks and two or three stores north of the Fuchs store, and was close to the “Mill Creek Sewer,” where the company had large iron tanks for storing oils, from which they filled sheet-iron wagons for distributing oils to retail dealers in the city, etc., etc.
The floor of the cellar of the Fuchs house was composed of a layer of two or three inches of cinders with a cement top constructed on the arch of the sewer. The sewer was fourteen feet wide, twelve feet high and with walls twenty to twenty-four inches thick. At the time of the explosion the river was very high and filled the cellars and first floors of the buildings on the levee. A fire occurred at the defendant company’s works on July 22, four days prior to the explosion. The witness giving the foregoing testimony, was Dr. Fuchs, a son of the deceased.
On his cross-examination this witness stated: (Generally ordinary sewage is dark and greasy looking; that he could not tell whether any petroleum oil was mixed with the water in the sewer, but that it had an odor like gas, not like ordinary lighting gas, but a greasy smell like petroleum or gasoline, something like that, though he was not sufficiently versed in chemistry to tell what kind it was; that the smell was not like that emanating from the black liquid which he had seen taken from sewers; that the smell was different from the ordinary gases from the gas works; that the manhole at the intersection of Broadway and- Fourth *186street was about one hundred to one hundred and fifty-feet west of the store; that he knew of no manholes in the sewer east of Fourth street; that after the explosion he saw flat ears which had fallen into the opening of the sewer which was constructed under the property before his father bought it, or built on it, and since that, time no repairs on the sewer east of Broadway had been necessary; that there was a slight current to the water in the sewer, the day of the accident, but the mouth of the sewer was blocked up by the river; that there was only one sewer inlet at the north end of Market at the junction of Fourth and Fifth streets, and one at the southwest corner of Broadway and Chouteau avenue, that the one at the north end of the Market was reconstructed, and it was made of clay pipe with a goose neck to it, to prevent the escape of gases into the open air; this was done at the instance of the people in the neighborhood who complained of the gases and odors thus escaping; that the inlet at the corner of Chouteau avenue and Broadway is intended to drain the surface water from the streets.
Follenius, whose marble works were located at 508. and 510 Chouteau avenue, and who had occupied those premises for about twenty-two years, and who had been familiar with the locality for some twenty-eight, years, testified that he remembered the fire which occurred at the oil company’s works on Thursday the twenty-second of July. There was a manhole at the corner of Broadway and Chouteau avenue on the west side of the latter. That on Sunday (next before the Monday the twenty-sixth of July) on which the explosion occurred, his place having connection with the sewer, he observed a peculiar smell from the sewer; that it seemed as though mixed with sewer gas and coal oil, which was different from the sewer gas smells which were there most all the time; that he was seated *187at his desk when the explosion occurred, and after that went out into his yard which looked as if it had been plowed up with a large plow; that large holes were blown in the top of the sewer three or four feet over, from which issued an odor like he had noticed the day before; that the shock threw the lid off the manhole at the corner and on going eastward he noticed the same kind of odor issuing at the southeast corner of Broadway and Chouteau avenue, also east of the Fuchs building, and that the smell was unlike coal gas and different from any smell that had come from the sewer before; that the odors that came from the manhole at the southeast corner of Broadway and Chouteau, were there before the sewer was built, but since its building could only be detected on a cloudy, rainy day. He further testified that he saw a colored mass mixed with the dirty water that was flowing down the sewer, of a kind of violet color, but could not say whether it was oil.
Tunstal, another witness, was in the back room of the building when the explosion took place, and Fuchs and Kriebaum were in there with him, and by that explosion he was cast into the sewer, and from which he got out after struggling in the stream a distance of about eight hundred feet; there he encountered water from five to six or seven or eight feet deep; that in the sewer were “sawdust and muck, and petroleum and coal oil, and everything else that you could think of that was nasty;” there was a little current, enough to carry him along, but none right where the debris from the house dammed up the sewer; that there was a large amount of sawdust in solution and general muck, more like molasses or tar, or something of that kind; that petroleum is a thick fluid like tar; and that there was coal oil on top of the water; this he recognized in the darkness of the sewer; that there was "gas either from petroleum or whatever it was that blew up the sewer;” *188“that the explosion didn’t ignite all the gas in the sewer; that there were other odors there besides that of gas; that witness was compelled to hold his nose and get under the water to keep from being asphyxiated with gas, which was just like needles going up into his nose;” that the defendant oil company’s works were located along side of the Mill Creek sewer, between Twelfth and Fourteenth streets on the Pacific railroad, and this was the only oil mill on that sewer.
Two hours after the occurrence of the explosion, another witness, Dr. Bowler, who was in the second story of the Fuchs building when it occurred, states that a characteristic pungent smell was noticeable arising from the opening in the sewer, such as usually arises from sewers; that after the explosion there was smoke or vapor arising from the sewer indicating that there had been fire or combustion in the sewer and that the gas had been consumed; that the ordinary gas which accumulates in sewers from decaying vegetable matter is explosive when ignited under certain conditions; that the density of the gas would have to be sufficient and also sufficient heat to produce the explosion ; that if all the conditions were favorable you would get spontaneous combustion or explosion by ignition, that is, applying a light to the accumulated gas; that sewer gas may explode under certain conditions without the presence of petroleum; that he understood the process of manufacturing gas from crude petroleum, which, and coal oil, are explosive when subject to a certain degree of heat, and that under favorable conditions either of the gases mentioned might explode spontaneously.
Schneider, who lived in the next door south of Fuchs, testified that there were two manholes in the Mill Creek sewer at the junction of Fourth and Fifth *189streets (the latter of which is usually called Broadway), one in the middle of the street and the other on the sidewalk on the west side of the street, and they were kept closed with solid covers; that there were two other openings to the sewer, one on the corner and one right in front of the Market, which were made for water to flow into, and two openings of usual size to the sewer also open, covered with grates in the alley, one hundred and forty-four feet east of the Euchs property, right behind the property of witness; these openings in the alley are to a sewer connected with the main sewer; that on the evening of the day of the fire (which occurred just a little after noon) the smell from the sewer of coal oil was so strong in his house that they had to close the windows; that his sewer connects with the Mill Creek sewer; that each succeeding night it was the same way, and they could not stand it in the rear of the house and had to go to the front because of this smell of coal oil. There was always some odor coming from the sewer, but not so strong as at the time mentioned; that the odor was much stronger after the fire than before; that since the manholes on Fourth and Fifth streets had covered tops they did not suffer se much from the odor; that the branch sewers affording connection from his house with the Mill Creek sewer had no self-acting safety cocks to shut off odors from the sewer in the yard.
Kuntz, a plumber, testified that seven or eight years ago he connected the premises of the Waters-Pierce Oil Company with the Mill Creek sewer at a point on Gratiot street, opposite Thirteenth, to drain their yard; that there was a grating at the junction of Fourth and Fifth streets over the main sewer to let the water flow in.
Hartung, reporter, testified that .he was present at. the time the fire of the oil company’s works occurred *190and was there some three hours; that he saw men who wore overalls and were laborers, who were acting under the command of Lindsay, chief of the fire department, digging trenches in among the railroad tracks and thus conducting, the oil and water ('which was' in large quantities in the ground from five to twelve inches deep) to the public sewers; that these men at work were about fifteen feet north of the oil company’s plant; that it was impossible to tell the proportions of the flowing water and oil; that the men were not firemen; that while the works of the oil company were furiously burning, the oil from the works ran out on the ground and among the railroad tracks on which the cars were standing; that he asked Chief Lindsay “whether there was any danger from that oil that was scattered among the tracks catching fire and damaging the railroad property, and he said, no, he thought not, that the men were leading it into those inlets, and I heard him say ‘move along,’ or ‘is it hot,’ and questions of that sort;” that there was no way to prevent the oil, etc., from escaping and going northward into the Union Depot yards where there were cars standing, except by turning it into the sewers; that the fire department was playing on the fire at the time the works were burning; that one of the officers of the defendant company stated to him that there was naptha in some of the company’s tanks. This witness also testified that there were three hundred or four hundred gallons of oil flowed into the sewers, and then after that three thousand or four thousand gallons, and then said he could not estimate the quantity as he could not tell how thick the oil was on the surface of the water.
■ Wilson, a policeman, testified that just as he passed the Euchs place the explosion occurred; that a dense smoke came out of the building which resembled the *191scent which arises after the extinguishment of a gasoline stove.
Enger, a gas engineer, who had twenty-one years experience in making gas from petroleum and its products, had not studied the matter of how gases originate that are formed in a sewer; that gases arise by ■evaporation; they are not made that way, but are produced that way, by simple evaporation; that some of the constituents of the oil would evaporate; that this was the case with all kinds of oil from crude petroleum down to naptha, gasoline; that illuminating oil that is sold as kerosene would not give off any vapor under ordinary temperature, nor would it produce any gas; “I have tried that many times;” that if you poured such oil on the floor, with a lighted candle in the room, the oil would not take fire; that it only explodes or burns by contact; that crude petroleum gives off a vapor under ordinary temperature, that is, any temperature of over sixty degrees Fahrenheit, which is inflammable by coming in contact with fire; that it can not be called a gas but is a vapor; that if you filled a large vessel with crude petroleum and put it in a closed room, after a time, depending on the atmosphere, there would be gas enough in the room to make it dangerous to go in there with a light; that there would have to be a great deal of ventilation in a room to prevent it being dangerous. “I would be afraid of it even in any case;” that he had given the question of the construction of sewers in a city no attention; that naptha is one of the first products of distillation of petroleum and is a very light oil, and vapors form from it much quicker than they would from petroleum alone; that gasoline is the same as naptha only a little lighter grade.
Asked whether sewer gas or marsh gas would not ignite or burn spontaneously, witness said that the *192former would not, and that ordinary gas would not to his knowledge; that it would take a very large quantity of crude petroleum to generate enough vapor to cause Mill Creek sewer to explode; that when he spoke of oil evaporating at a temperature of sixty degrees, he referred to oil in an open vessel, such an one as would admit of air; that from his general reading he knew that there was a constant generation of gases from animal and vegetable matter, and was called sewer gas, which is a mixture of gases; that there is a mixture of sulphuretted hydrogen and may be some marsh gas ;■ that marsh gas emanates from decaying vegetable matter; that witness believed it was common in sewers; that the conditions were favorable to it, and that the conditions were more favorable to sulphuretted hydrogen being in sewer gas and more so than ordinary marsh gas; that gas is generated in sewers from human excrements, rotting vegetables and animal matter; that explosions from sulphuretted hydrogen or marsh, gas would be about the same in their violence as gas from petroleum vapors; that in the opinion of witness if the explosion had resulted from crude petroleum or other cause, such sudden ignition would have raised the temperature to about two thousand, two hundred degrees, and the heat would have ignited and burned any petroleum in the sewer at the time, and would have ignited the petroleum more readily on the surface of the water, than if lying on a dry surface. This witness could not state what the temperature of the sewer was; that tank cars containing hundreds of barrels of oil are constantly being transported all over the country without any protection from the sun, and having on the top of each, a cupola top with a manhole in it and by ineans of such holes the cars are filled.
Humpert, who did business for Peters at French Market, across the street from Fuchs’ place, testified *193that on the twenty-sixth of July he went down in the afternoon' with a lighted candle to put some watermelons away in an ice chest in the cellar, when a cloud of fire came right in his face and knocked him on his knees, etc. This place was about one hundred and fifty feet from Fuchs’ place; and that there was a sewer in the cellar which connected with that one in the alley; that there was a manhole in the alley over the sewer and is covered by a grate; this was just back of Peters’ building.
Kern next testified that he was in front of Peters’' place when the explosion happened, that he saw Hum-pert, who had just come from the cellar with his clothing on fire; that witness took Humpert’s shirt off, the fire from his clothes, and in about the time a person could count ten, a loud report was heard and the front, of Fuchs’ building came out.
At the conclusion of the testimony the court, at the instance of defendants, gave instructions in the nature of demurrers to the evidence, whereupon plaintiff took a nonsuit, etc.
1. It will have been inferred from the foregoing quotations from the pleadings and the evidence that this cause requires consideration from two points of view, one relative to the pleadings, the other to the evidence.
In the first place, there is no evidence to show that the city contracted with the grantors and their assigns of Fuchs to “keep and maintain said sewer in good order and to care for the said sewer, so that said lot and any improvements which might be put thereon, would be free from danger of injury from and on account of said sewer and the use thereof.” This being the case, there is no right arising out of contract which could hold the city liable in the premises.
*1942. And it is patent of record that the other portions of the petition do not state that it was the duty of the city to keep Mill Creek sewer free from noxious or dangerous gases or free from fluids and substances which would generate such gases. Unless the duty of the city to do this is alleged in the petition, it states no negligence; for duty unperformed is the sole predicate of negligence, and without it, the latter can not exist. Cooley, Torts [2 Ed.], 791, 792; Railroad v. Stark, 38 Mich. 714; Cole v. McKey, 66 Wis. 500; 1 Shearm. & Redf. Neg. [4 Ed.], sec. 8; Hallikan v. Railroad, 71 Mo. 113.
The petition, therefore, states no facts sufficient to constitute a cause of action; a fatal defect which may be noticed in this court for the first time. Smith v. Burrus, 106 Mo. loc. cit. 97, and cases cited. Or, on which account objection could have been taken in the lower court to the introduction of any evidence. Butler v. Lawson, 72 Mo. 227.
Other matters in regard to the petition will receive comment in a subsequent paragraph.
3. Inasmuch as the trial court granted instruetions-in the nature of a demurrer to the evidence, it has been thought proper to make exhibition and profert of that evidence somewhat at large. In cases of this sort, as must be obvious, facts are mdispensable factors in determining the correctness of the action of the trial court in nonsuiting the plaintiff, since those facts must constitute the ¶<>0 arm of plaintiff’s action and of the defendant’s defense.
From the facts in evidence it appears illuminating oil that is sold as kerosene (coal oil) will not give off vapor nor produce gas under ordinary temperature; that it only explodes or burns by contact; that crude petroleum if placed in a large open .vessel in a closed room, would after a short time, if subjected to a tern*195perature of about sixty degrees Fahrenheit, give off sufficient vapor not gas, to cause an explosion if the room were entered with a light.
There was no evidence, however, as to what the temperature of the sewer was,' nor as to what the effect would be in the way of generating gas or vapor in a sewer where, according to the testimony, the proportion of the crude petroleum, etc., must have been exceedingly small when contrasted with the vast quantities of water contained in a sewer sixteen feet wide, twelve feet high, and from five to six, seven, and eight feet deep, even if we adopt the tare conjecture that there was as much as three thousand or four thousand gallons of oil turned into the sewer. It is true that the testimony shows that naptha, etc., would give off vapor at a much lower temperature than crude petroleum, but there is no testimony showing what the temperature of the sewer was, nor that any naptha, etc., was turned into the sewer on the day of the fire. So.that, under the testimony, we must put out of view as constituents of the litigated injury, naptha and gasoline, because not shown to have escaped from the tanks nor to have been conducted into the trenches leading into the sewer, and besides, conceding such escape and such conducting of those fluids, no temperature of the seiver shown. So that, under the testimony, kerosene or coal oil and crude petroleum must also be excluded from consideration as injury-producing ingredients, because the former does not generate either gas or vapor under ordinary temperature — sixty degrees —nor the latter generate anything but vapor, not gas, under that temperature, and no testimony as to what degree of heat or cold existed in the sewer, and gases, not vapors are alleged in the petition as the cause of the explosion. These things alone would certainly seem *196to warrant the ruling of the trial court in giving the instructions complained of.
4. But other inferences are to be drawn from the facts in evidence already related which, if possible, even more strongly tend to support the conclusion, reached by that court. It can not be known with any plausible degree of probability from the facts developed in evidence what was the cause of the explosion. No-one can carefully read the testimony, and, after due-deliberation upon it, be enabled to say what gas or combination or commingling of gases produced the-unfortunate result which gave origin to this action.. The conditions were favorable, as the evidence shows,, to the generation of several gases, viz.: methane or marsh gas or carburetted hydrogen, formed by the decomposition of vegetable matter under water, also known as fire damp, colorless and inodorous, which is the cause of the explosions which so frequently take-place in coal mines, and is given oft when the mud ia stagnant pools and marshes is stirred; and the indications are of the possibility of making this gas from the elements since its constituents may be thus formed (Eemsen, Organic .Ghem. 23, et seq.j and which constitutes the most abundant ingredient of coal gas; (Fownes, Elem. Ohem. 299) sulphuretted hydrogen, or hydrogen sulphide, also a colorless gas, but by no-means inodorous, having the odor of putrid eggs, and being the frequent product of the putrefaction of organic matter, both animal and vegetable. Ib. 173.
Now, if we say nothing of-a gas or gases which might result from an admixture of those aforesaid, and if we admit that the conditions were also favorable to the generation of gas in the sewer from the oils introduced therein, as one of the incidents of the fire, still we are confronted by the rule which declares that where an action is brought for damages, which are *197•occasioned by one of two causes, for one of which defendant is responsible, and for the other .not, the plaintiff is fated to failure if his evidence fails to show that the damages were produced by the former, or if, from the evidence, the probabilities are equally strong that the damages were caused by the one as by the other. Searles v. Railroad, 101 N. Y. 661. This principle finds recognition in Priest v. Nichols, 116 Mass. 401, and Smith v. Bank, 99 Ib. 605.
5. Eecurring for a moment to the petition, preparatory to a further discussion of the evidence from other points of view, we find that it charges that “said ■sewer was provided with openings especially designed to carry off any gases which might arise in said sewer, and be liable to combustion and explosion, etc., etc., yet said city, its agents and servants, knowing that «aid defendant the Waters-Pierce Oil Company, had flooded said sewer with oil, neglected to open said vents [and carelessly and negligently to take measures and precautions to prevent gases arising in said sewer .so as to endanger the same,]” etc.
The words not included within the brackets are those which allege plaintiff’s cause of action, because where a particular act of negligence is specified as a cause of action, there evidence will not be received to support a general allegation of negligence, but the plaintiff will be confined to the act of negligence specifically assigned. Schneider v. Railroad, 75 Mo. 295; Waldhier v. Railroad, 71 Mo. 514.
From which premise it results that no evidence was properly admissible in regard to the words in brackets. Besides, those words were but the statement of a legal ■iconclusion; something not traversable; no issue of fact could be raised upon them. Bliss, Code Plead. [3 Ed.], •secs. 212, 213, 413. Under our code, the facts in pleading are constitutive, and in order to be proved *198must be distinctly alleged. Pier v. Heinrichoffen, 52 Mo. 333; Bank v. Hatch, 78 Mo. 13; McKinzie v. Mathews, 59 Mo. 99; Nichols v. Larkin, 79 Mo. 261; Lanitz v. King, 93 Mo. 513.
Taking, then, the facts specifically assigned as negligence, and contrasting them with those offered in evidence in their support, we find that within a radius-of two hundred and fifty feet from Fuchs’ place of business there were four openings through which the gases in the sewer could escape, saying nothing of the sewer connections at Peters’ store, and at Follenius marble works, and so the only thing that remains of' plaintiff’s claim of the city’s negligence in this regard is as to the manhole in the center of the street where Fourth and Fifth streets intersect each other, and that, at the sidewalk on the west side of Fifth street.
Respecting the first one, Dr. Fuchs’ testimony shows that it was constructed with a goose neck so as t-o prevent the escape of gases, having been changed from a straight pipe, because the people in the locality complained of the odors formerly coming from it, so-that even if the covering had been removed from this manhole, no gases could have escaped, and it will not be presumed that plaintiff intended to include in her petition this manhole, but only those whose covers if removed would have given ventilation to the sewer; that is to say, egress for the gases therein.
As to the second manhole cover, it was the only one which could have been removed that was not removed. But did the nonremoval of this one so retard or prevent the escape of gases as to cause the accident?' If it did, then the burden is on jplaintiff to show that it did. It devolved upon her to ‘ ‘prove facts and circumstances, from which it can be ascertained with reasonable certainty what particular precaution the defendant ought to have-taken but did not take” (1 Shearm. & *199Redf. Neg. [4 Ed.], sec. 57), which of course would include as a legitimate corollary therefrom, that had such particular precaution been taken, the reasonable probability is that the accident would not have occurred.
Thus in Daniel v. Metropolitan R’y Co., L. R. 3 C. P. 216, Willes, J., said: “It is necessary for the plaintiff to establish by evidence circumstances from which it may fairly be inferred that there is reasonable probability that the accident resulted from the want of some precaution which the defendants might and ought to have resorted to; and I go further, and say that the plaintiff should also show with reasonable certainty what particular precaution should have been taken.” Though the judgment in this ease was reversed on another ground, this doctrine was distinctly affirmed in the same volume (591), and in the house of lords (L. R. 5 H. L. 45), and the language employed by Willes, J., has been frequently cited and quoted with approval. Hayes v. Railroad, 111 U. S. 228; Railroad v. Stebbing, 62 Md. 504; Williams v. Railroad, L. R. 9 Exch. 157; Railroad v. Stout, 17 Wall. 657; Randall v. Railroad, 109 U. S. 478; Lovegrove v. Railroad, 16 C. B. N. S. 669.
In the case at bar there was no attempt to make the proof here indicated as necessary; in short, to connect the neglect to remove the cover of the single manhole with the accident.
In illustration of this principle it has been ruled where the jury are told that if all the evidence satisfied them that there had been negligence on the part of the defendant, although they might not be able to satisfy themselves in what that negligence consisted, they would be authorized to find a verdict for plaintiff, that such a charge was erroneous; that if the jury could not find any rational ground upon which to impute negli*200gence to defendant, they should give a verdict in its favor. McCaig v. Railroad, 8 Hun, 599; Searles v. Railroad, supra.
6. Again, if the city is to be held responsible for failing to keep open the vents to the sewers within its jurisdiction, is it to be held liable also if some person passing while the vents are open casts a lighted match into one of them, or the gas from it rises and catches fire from a street lamp, thereby, causing an explosion? Is it possible that the city be thus held responsible whether it does or does not open vents? And yet if the position taken by plaintiff as ground for recovery in this action be correct, that the city is responsible for the gases which breed in its sewers, then the spectacle will soon be presented of actions for damages against the city, because: First, it does not open its sewers and thereby allow the gases therefrom to escape, thereby causing an explosion; because, second, it does open its sewers, and thereby an explosion is caused; because, third, it opens its sewers to allow the gases to escape, and thereby becomes liable for disease and death scattered by reason of the escape of such gases; because, fourth, it does not pump out the sewage from the sewers, or at least does not use a liberal quantity of disinfectants so as to deodorize - the contents of the sewers and thus render them if not sweet, at least innocuous !
Such are the possibilities of municipal liability which present themselves if the present action can be maintained. And if it can, it might be well to suggest that if the city is thus to be made an insurer, it ought, at least, as some compensation, to be allowed to issue accident policies, and take premiums on the multitudinous risks it is thus compelled to assume. Hitherto it had been supposed that it was the peculiar and exclusive purpose and function of sewers, and that they were *201adapted, devised, and designed to conceal and cany off the foulness which accumulates where great bodies of people congregate, and not to disseminate mephitic odors and gases, thus poisoning the atmosphere throughout the city.
7. Furthermore, “Mill Creek Sewer,” as was conceded at the trial, was constructed in a manner that left nothing to be desired; it had been built some thirty-four years, and no accident of the nature now presented had ever occurred in it. Indeed, it does not appear that an occurrence ol such sort had ever before happened. Now it is settled by abundant authorities and by numerous and frequent adjudications that it is not negligence to omit a precaution which, if taken, would have prevented the injury, when the injury could not reasonably have been anticipated and would not, unless in exceptional circumstances, have happened, because of the omission. Such instances are assigned to the domain of inevitable accident, for which, no one being negligent, no one is responsible.
Thus in Dougan v. Champlain, etc., Co., 56 N. Y. 1: “D., plaintiff’s intestate, was a passenger upon defendant’s boat on Lake Champlain. The forward deck was surrounded by bulwarks three or four feet high, with gangways upon each side closed by rails hinged to the bulwarks and of the same height, and coming down upon stanchions in the center of the .gangway, leaving the space beneath open. This deck was not designed for passengers, but they were permitted to come upon it with knowledge of defendant’s employees. D. came out thereon, his hat blew off, he .sprang to recover it, slipped under the gangway rail, fell overboard and was drowned. It appeared that all the boats upon the lake were constructed in the same manner; that they had been so run for many years, and there was no proof tending to show that anyone *202had ever before gone overboard in this way, or that-such danger had been apprehended. Held, that the evidence failed to show negligence on the part. of defendant, and that plaintiff was properly nonsuited.”
So, too, in Hubbell v. Yonkers “plaintiff was riding along one of defendant’s streets, the roadbed of which was thirty feet wide, macadamized and in good condition. On one side, where the street was graded up-about twelve feet, there was a sidewalk ten feet wide, separated from the roadbed by a curbstone eight inches high. There was no fence, wall, or other obstruction to guard the outer edge of the sidewalk. The horse-attached to the wagon in which plaintiff was riding became frightened and commenced to shy, and, in spite of the efforts of the driver, went over the curbstone and sidewalk and down the embankment, carrying the wagon and plaintiff with him. In an action to-recover damages, for injuries received by plaintiff, it appeared that the street had been in the same condition since its opening, over ten years before, and, so far as appeared, no similar accident had occurred. Held, that defendant was not liable, that the accident was one of a class so rare, unexpected, and unforeseen, defendant could not be charged with negligence for a failure to guard against it.” 104 N. Y. 434.
A mule caught its foot in a hole in a railroad track so small that no one could have foreseen such result. Held, no liability. Nelson v. Railroad, 30 Minn. 74.
Similar nonliability was announced where a workman was painting by lamplight the inside of a tank with an approved and long used paint, bought ready for use, and the benzine in the paint caused an explosion. Allison Mfg. Co. v. McCormick, 12 Atl. Rep. 273.
From some unexplained cause a telegraph wire across a track sagged, and hitting a brakeman on top *203of a car, broke, at the same time becoming fastened to the car brake. The end caught a man engaged in business near the depot, and the wire drawn along by the moving train the man was killed. Held, to be an accident. “Negligence,” says Mitchell, C. J., “is not to be presumed upon the fact of an occurrence like that involved in the present case, the statement of which suggests its anomalous, exceptional, and extraordinary character.” Railroad v. Locke, 14 N. E. Rep. 391.
Like rulings have been announced where accidents have happened from machinery, where their liability to happen is proved only by their actual happening. Richards v. Rough, 53 Mich. 212; Sjogren v. Hall, Ib. 274.
In O'Mally v. Railroad, 113 Mo. 319, the tunnel had been used for thirteen years, and in an action brought for the death of the plaintiff’s husband, employed in the defendant’s tunnel, through which it operated locomotives and cars, and the petition charged that the tunnel, because the fan that ventilated it was-out of repair, was in a dangerous condition, being filled with steam, smoke, and poisonous gases; and that defendant, well knowing this fact, which was unknown to the deceased, negligently ordered him to go into the tunnel, whereby he was choked, strangled, and killed. Held, that, as there was total failure of the evidence to show that the smoke in the tunnel when decedent entered was dangerous to human life, or to show that defendant could have anticipated a condition of the tunnel dangerous to human life, plaintiff could not recover. To the like effect see Cooley on Torts, 91, et seq.; Withers v. Railroad, 27 L. J. Exch. 417; Loftus v. Ferry Co., 84 N. Y. 455; Cleveland v. Steamboat Co., 68 Id. 306; Sutton v. Railroad, 66 Id. 243; Bishop, Noncont. Law, secs. 182, 447; Bishop v. *204Railroad, 14 R. I. 314; Wright v. Wilmington, 92 N. C. 156.
The same principle is recognized in Flori v. St. Louis, 69 Mo. 341, where the city was held liable to a person for injuries inflicted by the fall of a market house caused by a windstorm of unprecedented force and violence. It is unnecessary to say here whether the case might not have rested on another ground. It is certainly opposed, in any event, to a recovery by plaintiff.
8. Moreover, the defendant city in the construction of “Mill Creek Sewer” and in its maintenance, was and is engaged as a governmental agency in the performance of a public, sanitary duty for the public good, and not for its own private advantage or •emolument. In such circumstances, it is well settled in this state, as well as in many other jurisdictions, that a municipality is not liable in damages for the wrongful or negligent acts of its officers and servants, unless made thus liable by positive law or by inevitable implication. Murtaugh v. City, 44 Mo. 479; Heller v. Sedalia, 53 Mo. 159; McKenna v. St. Louis, 6 Mo. App. 320; Armstrong v. Brunswick, 79 Mo. 319; Carrington v. St. Louis, 89 Mo. 212; Maxmilian v. Mayor, etc., 62 N. Y. 160; Hill v. Boston, 122 Mass. 344; Detroit v. Blackeby, 21 Mich. 84; 2 Dillon, Mun. Corp. [4 Ed.], secs. 965, 965a, 975, 976, 977, 980, and cases cited.
Carrington v. St. Louis, supra, while it correctly states the principle applicable to this class of cases, yet its application in that instance suggests an interrogation point, as to which see sections 58, 60, 210, 974, 975, Dillon, Mun. Corp. [4 Ed.].
Besides, if the theory contended for by plaintiff is to prevail, it would result in casting on defendant city a task impossible of performance, as already stated, *205and one which if it conld be performed would subject the city to fresh liabilities by reason of such performance, and in addition thereto, would defeat and destroy the very purpose and function which a seioer is obviously designed to accomplish, to ivit, to prevent the air of the municipality from being contaminated by foul odors, and other contagious and infectious gases and emanations.
"When a city has, as in this instance, built a sewer in a most admirable manner and has kept such health-preserving conduit free from obstructions, its complete duty, whether considered a public or a corporate one? has been entirely discharged.
9. It only remains to say that there is nothing in the facts in evidence which by any possibility casts any blame or liability on defendant oil company. It can' not be considered as having permitted the oils to escape and run into the sewer, merely because it did not forbid the oils which ran from its premises into the streets and on the railroad tracks from being turned, by means of trenches dug into the sewer, nor because it did not use force to prevent this from being done.
For the foregoing reasons the judgment should be affirmed; and for which reasons I dissent from the majority opinion.
Buegess and Robinson, JJ., concur.