In this action by Earl McGaugh against the City of Fulton, for personal injuries resulting from an explosion, the court sustained the city's motion for judgment in accordance with its previous motion for a directed verdict after a jury had been unable to agree upon a verdict. The court sustained the motion upon the specific grounds that the plaintiff's evidence failed to establish a cause of action against the city or to show any causal connection between the alleged negligence of the city and the explosion and injuries. Upon this appeal by McGaugh the sole question then is whether his evidence reasonably demonstrates negligence on the part of the city, causal connection, and consequent prima facie liability.
McGaugh and his family were tenants in a four-room frame house at Eighth and Walnut Streets. The house had a full concrete basement with four windows which were closed at the time of the explosion. Entry to the basement was by means of a trap door on the back porch. The McGaughs had lived in the house ten days when Earl went to the basement about eight-thirty on the evening of January 21st, 1943 with a lighted cigarette in his mouth and, immediately upon his entering the basement, there was a terrific explosion which injured him and damaged the house. The house was raised off the foundation and the floors collapsed. No one had previously detected an odor of gas in the basement and once or twice prior to the explosion the McGaughs had done the family washing in the basement, using a three-burner coal oil stove which they carried to the basement from the kitchen.
It was the plaintiff's theory that the explosion was of sewer gas or natural gas or a combination of both sewer gas and natural gas. The plaintiff's amended petition, as it stood upon the trial of the cause, charged negligence on the part of the city in the construction, *Page 1127 maintenance and operation of its municipally owned gas and sewer system so as to permit gas to escape into the plaintiff's basement. It was alleged that the city was negligent in the installation of its pipes, in failing to find and repair a break in its natural gas main about 400 feet from the plaintiff's residence, in the construction and maintenance of its sewer main on Seventh Street with an unvented dead end and in failing to keep the vents in its manhole covers open along Seventh and Walnut Streets. When the case was submitted to the jury the plaintiff proffered instructions hypothesizing the city's negligence in connection with its sewer system but did not proffer an instruction hypothesizing negligence concerning natural gas. Nevertheless the plaintiff now claims that his evidence demonstrated negligence on the part of the city in connection with both its sewer and natural gas systems.
The house was built in 1940 and all sewer installations and appliances were installed by a plumber under contract with the owner and builder. The plumbing consisted of a lavatory, stool, bathtub and kitchen sink on the first floor. These were all connected [549] with individual traps to a soil pipe which passed through the floor of the basement and were connected with the city sewer at the property line on Walnut Street. Between the center of the basement and its west wall there was a drain in the floor covered by a grill and set in a sump in the floor. The drain was equipped with a trap and entered the sewer line with a "Y" connection. Between the drain and the west wall there was a cleanout pipe. As we understand the record the openings in the basement plumbing were, first, the cleanout pipe, second, the floor trap which connects with the "Y" and, third, was a vent or stack from the bathroom up through the house. The inference, as we understand, was that the cleanout pipe should have been connected with the sewer line ahead of the "Y" rather than back of it. It should be interpolated that the record is not perfectly clear whether the stack through the roof was connected with the sewer back of the plumbing fixtures or whether it was ahead of the fixtures only. But, whether these two latter points are clear or not, the cleanout pipe was uncapped at the time of the explosion and had been for some time, we assume for at least ten days. McGaugh saw the pipe but did not know its purpose or that it was supposed to have been capped. He supposed it was a part of the drain.
The city sewer line extended from Seventh Street up Walnut Street about 400 feet and terminated in a dead end almost in front of the McGaugh house. There was no manhole at the dead end sewer, but there was a lamphole of tile covered with an iron cap. There was a manhole at the intersection of Seventh and Walnut Streets, another 300 feet west of the intersection on Seventh Street and one 200 feet east of the intersection. The evidence was that all the manhole covers were perforated or contained holes but the holes were all closed up *Page 1128 with asphalt paving and had been for some time. The city did not offer any evidence but one of its employees, testifying for the plaintiff, said that manholes were for the purpose of permitting work on the sewer lines and that the theory of holes in manhole covers for the purpose of venting sewers had been discarded. However, there was positive testimony from another witness that the holes in manhole covers served the dual purpose of providing access to them and of venting them, to allow the "escape of air and gas from the sewer and also allow the introduction of fresh air into the sewers." There was testimony that sewer gas accumulated in the sewers and that venting them through manhole covers permitted the gas to escape. The composition and explosive nature of sewer gas was described. Two witnesses, both city employees, testified that the uncapped cleanout pipe in the basement drain would permit gas to enter the basement. One of these witnesses, the superintendent of the gas department, testified that shortly after the explosion he examined the premises and found the uncapped cleanout pipe. He said, "I found a combustible gas in the sewer. Inside the sewer. In the cleanout opening." He used a Davis Emergency Combustible Indicator and it registered "Dangerous." He was unable to separate the kinds of combustible gases but, he said, "There was a trace of sewer gas."
"Q. Was there any trace of natural gas?
A. There was a trace of combustible gas. I couldn't sort them out at the time.
Q. Then it could have been either one as far as you are concerned?
A. Yes, could have been.
Q. It could have been both?
A. It could have been something else."
The natural gas line ran in front of the McGaugh residence along Walnut Street between Seventh and Eighth Streets. The gas line was laid about five feet and eight inches west of the east sidewalk in the parkway. As we understand the record, the gas lines did not intersect the sewer lines but, as the appellant says, the two lines paralleled one another for some distance up Walnut Street. Nor, as we understand it, was the McGaugh house connected with the natural gas main. But, on March 18, 1943, fifty-eight days after the explosion a complete break in the gas main was found. The break was in the Seventh Street main about twenty feet from the intersection of the gas line on Walnut Street and approximately 500 feet from the McGaugh residence. The city employee who found and [550] repaired the gas break testified that in his judgment the break was not over a week old. He said that the ends of the pipe were spread apart about one-quarter of an inch and were bright. There was no corrosion on the pipes at the break and the earth around the pipes was dried out eight or nine inches in each direction from the pipes. *Page 1129
It is the position of the appellant that these circumstances demonstrate negligence on the part of the city and constitute substantial evidence that the explosion was of sewer gas, natural gas or a combination of both.
[1] Clearly, in the circumstances detailed, there was no evidence, either direct or circumstantial, from which a jury could reasonably find the city negligent as to its natural gas system. Here, as to natural gas, a jury could not reason upon the facts established by the evidence and infer from them, except by mere conjecture and guess, the further necessary facts demonstrating a breach of duty, negligence and consequent liability, in the maintenance and operation of the city's natural gas system (Cf. Hardwick v. Kansas City Gas Co., 355 Mo. 100,195 S.W.2d 504, 508) or, that any negligent conduct respecting natural gas contributed to cause the explosion and injuries complained of. As the appellant says, it was impossible to determine how long the break in the gas main had been there. It may be that the employee who found the break was not qualified to express opinions on the subject, nevertheless his is all the evidence there was on the subject and that indicates that the break was less than a week old. There was but slight discoloration of the earth around the pipe and the break was bright, giving the appearance of being recent. Cf. Stephens v. Kansas City Gas Co., 354 Mo. 835, 191 S.W.2d 601 and Hardwick v. Kansas City Gas Co., supra. There was no evidence here of a break in the gas line and the escape of gas prior to the explosion and, hence, the plaintiff did not prima facie demonstrate negligence as was the case in Nomath Hotel Co. v. Kansas City Gas Co., 204 Mo. App. 214, 223 S.W. 975,253 S.W. 975; Sipple v. Laclede Gaslight Co., 125 Mo. App. 81, 102 S.W. 608; Vaughn v. Kansas City Gas Co., 236 Mo. App. 669,159 S.W.2d 690 and Taylor v. St. Joseph Gas Co., 185 Mo. App. 537, 172 S.W. 624. The appellant urges, due to the well known propensity of gas to percolate through the earth into openings and cavities, that it is not necessary for some witness to positively and unequivocally testify that natural gas from the broken main entered his basement. But, the break in the gas main was not discovered until fifty-eight days after the explosion and, if it could be found that natural gas was involved in the explosion, there is no fact or circumstance from which it could reasonably be found that it came from this broken main. Furthermore, in this case, the possible course of natural gas from this or any other broken main to the sewer line, to the plaintiff's basement and its probability as a cause of the explosion was not clearly shown as it was in the Stephens and Hardwick cases. See also and compare: James v. Kansas City Gas Co., 325 Mo. 1054,30 S.W.2d 118 and Nomath Hotel Co. v. Kansas City Gas Co., supra. In short, there was no substantial, probative evidence from which a jury could reasonable find that the city had been negligent with respect to its natural *Page 1130 gas system (Nomath Hotel Co. v. Kansas City Gas Co., 204 Mo. App., l.c. 232, 223 S.W., l.c. 982) or that any alleged negligence concerning it contributed to cause the explosion and injuries complained of (Messmer v. St. Louis County Gas Co. (Mo. App.), 42 S.W.2d 963, 965) and, hence, the court did not err in refusing to submit this issue to the jury.
[2] It is in this connection that the propriety of the plaintiff's hypothetical question to his expert witness and its admissibility becomes important. The question, among other things, hypothesized "a break in the gas main . . . approximately five hundred feet . . . from the McGaugh residence." As we have indicated, and assuming the question to be otherwise proper, this part of the question was not based upon an assumption of fact from which negligence could be inferred and the fact did not reasonably permit the inference of causal connection between the break in the gas main and the explosion. [551] Compare Golden v. National Utilities Co., 356 Mo. 84, 201 S.W.2d 292 and Stephens v. Kansas City Gas Co., supra. Furthermore, upon objection to his hypothetical question the plaintiff did not call upon the defendant to point out in what respects the question was objectionable as not being based upon the evidence (Crockett v. Kansas City Rys. Co. (Mo.), 243 S.W. 902) nor did he offer to amend the question so as to eliminate any facts not in evidence or to supply any omission in the question (Marshall Livery Co. v. McKelvy, 55 Mo. App. 240) and so, there being no basis for the assumption, the question as put to the witness was improper and it was not error to exclude it. Henson v. St. Louis-S.F. Ry. Co.,301 Mo. 415, 256 S.W. 771; 3 Jones, Commentaries On Evidence, Sec. 1326, p. 2425.
[3] As the appellant points out, the unquestionable fact was that some kind of gas exploded in the plaintiff's basement. But it does not follow that the city is liable for the explosion unless there is proof of negligence and of facts and circumstances from which the jury could reasonably find that the city's negligence with respect to its sewer system caused the explosion. Brauer v. St. Louis County Gas Co. (Mo. App.), 238 S.W. 519, 521. As it was with natural gas so it is with sewer gas, the plaintiff's evidence must establish with reasonable certainty facts upon which the jury could exercise its function of reasoning and thereby draw the inference not only that the city was negligent with respect to its sewer system but that its negligence respecting the sewer system caused the explosion and injuries complained of James v. Kansas City Gas Co.,325 Mo. 1054, 1069, 30 S.W.2d 118, 123; Hardwick v. Kansas City Gas Co., supra.
[4] In this connection it may be assumed, for the purposes of this opinion, that the city ordinances with reference to the installation of plumbing in residences were admissible in evidence, not as demonstrating liability but as a circumstance bearing on its duty and what it might be expected to anticipate (Bean v. City of Moberly, *Page 1131 350 Mo. 975, 169 S.W.2d 393; Ryan v. Kansas City, 232 Mo. 471, 134 S.W. 566) even though there may have been no duty upon the city to inspect the appliances in the house and even though it may not have been liable for any negligence with respect to the plumbing appliances in the house. Barrickman v. National Utilities Co. (Mo. App.), 191 S.W.2d 265. While liability for negligence may not be predicated upon hindsight, nevertheless the city may have been bound to anticipate, if it were negligent with respect to its sewers and any gas in them, that such negligence might result in gas escaping from its mains into a basement through improper house installations or through an uncapped cleanout pipe. Guthrie v. City of St. Charles, 347 Mo. 1175, 1190, 152 S.W.2d 91, 98. We may also assume, for the purposes of this opinion, that it is a matter of general knowledge that sanitary sewers contain gas, some of which is explosive when mixed with air and exposed to some kind of ignition. Guthrie v. City of St. Charles, 347 Mo. l.c. 1189, 152 S.W.2d l.c. 98. It should be interpolated, as to explosive sewer gas, however that this categorical statement — as a universal truth or as a matter of common knowledge — may be subject to some doubt or qualification at least in the absence of evidence concerning the entire sewer system, its construction, the manner of its operation and maintenance or some extraordinary fact or circumstance specifically accounting for the presence of or accumulation of explosive quantities of sewer gas in a particular section of a sewer system.
[5] The plaintiff's expert witness testified that "sewer gases" naturally and normally accumulate in sewers and that the purpose of venting manhole covers is to permit the escape of sewer gas and the introduction of fresh air into the sewers. In this case there was no evidence concerning the city's sewer system as a whole, its construction, operation and maintenance. The evidence was confined to approximately 800 feet of the sewer system near the McGaughs' residence. But as to that section of the sewer the inference is plain, from the plaintiff's evidence, that it was improper to cover the manhole vents with asphalt and to permit the sewer to terminate in a dead end with an unvented lamphole. This fact, or the evidence from which it is a fair inference, when considered with the assumptions [552] we have previously made may support the inference that the city was negligent with respect to this section of its sewer line and particularly with respect to its unvented manhole covers and the unvented dead end sewer and yet it does not follow, in the circumstances of this case, that the plaintiff has demonstrated that the city's negligence in this respect caused the explosion and injuries complained of.
[6] As we have said, the city gas superintendent testified that he found a dangerous amount of "combustible gas" in the cleanout pipe in the basement. There was "a trace of sewer gas." It is not possible to say, with certainty, form the record whether he also found natural *Page 1132 gas in the basement sewer but, in any event, we have found that there was no evidence of negligence respecting natural gas. When asked whether there was any trace of natural gas he said, "There was a trace of combustible gas. I couldn't sort them out at the time." He said that it could have been either one or both and that "It could have been something else." Here the origin and accumulation of any explosive sewer gas and its probable course from the point of accumulation or confinement to the place of explosion is not traced with reasonable definiteness or accounted for as it was in Guthrie v. City of St. Charles and James v. Kansas City Gas Co., supra. In the Guthrie case all other kinds of gases were eliminated as a possible cause of the explosion. In that case there had been an odor of sewer gas for some time. There an old sewer line paralleled a new one and a manhole cover was left off of a manhole in the old sewer. The elevation and connection of the sewers was shown and it was demonstrated that water from an unprecedented rain entered the open manhole and pushed the gas ahead of it into the basement where it exploded. In fact the open manhole in the Guthrie case was alone sufficient evidence to require the submission of that case to the jury. Limpert v. City of St. Charles (Mo. App.), 158 S.W.2d 249. In the James case, as well as in the Stephens and Hardwick cases, the presence of explosive gas, other than natural gas, was reasonably accounted for as a probable and possible cause of the explosion.
But in this case, regardless of the tenuousness of any inferences on that subject, with the hypothetical question to the plaintiff's expert witness eliminated there is no evidence, as the question put it, "as to the potential cause of this explosion." Here we do not even have the choice of drawing inferences from the contrary opinions of experts as to the probable cause of the explosion as was the case in Hardwick v. Kansas City Gas Co., supra. Even assuming negligence in respect to the unvented manhole covers and the proof of "combustible gas" there is no evidence from which causal connection between the negligence and the explosion could be found without resorting to mere conjecture and speculation. Fuchs v. City of St. Louis,167 Mo. 620, 67 S.W. 610. It follows that the court was not in error in sustaining the motion for a directed verdict upon the specific grounds that the evidence, construed favorably to the plaintiff, failed to establish a cause of action against the city as to natural gas and failed to demonstrate or hypothesize causal connection between the negligence of the city and the explosion as to explosive sewer gas.
[7] Nevertheless, since it appears that the plaintiff may be able upon another trial to adduce evidence, hypothetical or otherwise, demonstrating causal connection, the judgment is reversed and the cause remanded. Finnegan v. Missouri Pac. R. Co., 244 Mo. 608, 149 S.W. 612, Knorp v. Thompson, 352 Mo. 44, 65-66, 175 S.W.2d 889, 902; Byrne v. Prudential Ins. Co. of America (Mo.), 88 S.W.2d 344, *Page 1133 347; Hummel v. American Mfg. Co. (Mo. App.), 279 S.W. 202, 204; Hill v. East St. Louis Cotton Oil Co., 202 Mo. App. 478, 489,214 S.W. 419, 422-423. Westhues, C., dissents; Bohling, C., concurs in result.