This action was instituted by appellant against respondents, the company as the seller and the other respondent as its agent. At the conclusion of all of the evidence on behalf of appellant, the trial court granted a motion by respondents for a nonsuit and dismissed the action.
Appellant is a nonprofit corporation operating a junior college at Bolling Bay on Bainbridge island. Prior to the opening of the school year of 1932-1933, it was decided to equip a science laboratory and classroom in the basement of one of its buildings near the beach. The basement floor was of concrete and sloped approximately a foot in the width of the building.
All of the construction work in equipping the laboratory was done by and under the direction of one Squance, maintenance man for appellant for about eighteen years. He qualified as a carpenter, plumber and heating man. To get a level floor, he put 4" x 12" joists in over the concrete floor. Notches were cut in these joists to accommodate the various pipes that would be used in the laboratory and classroom, such as hot and cold water, gas, air, and vacuum. Then a two-inch wooden floor was laid upon these joists and on top of that an inch and a half of concrete, which formed the floor of the laboratory and classroom. This floor was about six inches above the original basement floor at one side of the building and about eighteen inches at the other side.
The space between the two floors was enclosed on all sides by concrete walls, but there were two openings in the wall between the enclosed space and the furnace room. One was an eight-inch hole to permit water to drain into a sump in the furnace room and the *545other was a three-inch hole through which ran a two-inch pipe to carry away the water which a pump lifted from the sump. The pipe ran through the enclosed space between the floors and discharged the water at the other end of, and outside, the building.
The pump was operated by an electric motor, which went on and off automatically by means of a float in the sump. When the water reached a certain height, the float caused the pump to “kick on,” and when it dropped to a certain point it “kicked off.” The motor was going on and off about every fifteen minutes between November 4th and 9th, 1932.
William E. Ward was employed by the college as a teacher of high school and college chemistry for the school year of 1932-1933, and was also a teacher of other science subjects. He was a man of nine years’ experience, familiar with the uses of gases, knew their highly inflammable character and that escaping gas released into a chamber containing air formed a highly combustible and explosive mixture.
Ward testified that he had taught chemistry for eight years; that, when he arrived at Moran in the fall of 1932, the laboratory for the science course was not equipped, and it was part of his duty to select the equipment and apparatus and to choose the type of gas to be installed. While in Seattle, he went to the office of respondent the oil company, which manufactured Flamo gas. He explained their purposes and needs, and two or three days later respondent Allison came out and went over the building. The pipes for the gas had been installed before Ward arrived and the floor built over the pipes, and the end of the pipes came up through the floor to the outlets, where they would be connected with the Bunsen burners on the laboratory table.
*546Allison saw the pipes and installation that then existed, asked several questions about how they Were put in, the manner in which they were constructed, about which Ward knew nothing and referred him to Squance, the maintenance man, for information as to the installation of pipes and work of that nature. Squance explained how the pipes had been put in, and Allison said it would be all right. When asked his opinion about the size of the pipe, Allison said the usual size was larger than that, and that he would recommend larger pipes, but since it would be necessary to tear up the whole floor to put it in, that under the circumstances it would, perhaps, be better to leave the existing pipes; that he recommended half-inch pipe along the laboratory table, as that would give more pressure.
Squance told Allison that the pipes underneath the floor were all new and sealed with white lead. Squance freely admits that all Allison knew about the condition of the pipes was what Squance told him. Allison told Ward and Squance that the Flamo set must be installed outside of the building. The contract was then signed by appellant, calling for the delivery to it of a Standard Flamo dispensing set called “Flamo Equipment.” Appellant agreed with Allison that it would install an additional pipe from the Flamo set into the building to connect with the pipes rising from the floor in a closet. A half-inch pipe line was later installed by Squance, and a shut-off valve was placed in the closet between the new half-inch pipe and the quarter-inch pipe which ran under the floor of the basement.
In the contract signed by the parties is the following .clause:
“Company [Standard Oil Company of California] shall not be liable for any loss or damage to persons or property arising out of or in any way connected with the use and/or installation of said equipment and/or *547Flamo cylinders and/or Flamo, dne to leakage, fire, or other cause.”
On October 19,1932, the company delivered a Flamo set to appellant and it was installed on the north and outside of the building. The set consists of a metal cabinet containing two cylindrical tanks about four feet in height and ten to twelve inches in diameter, with an outlet on the top, valves and a cut-off to be connected with the pipe. On the set is a disc called a regulator and a gauge above that; there are valves on the top of each cylinder to turn the gas on and off. It was explained to Ward and Squance that the regulator reduced the pressure from high in the cylinder to low in the line. After the Flamo set had been installed, Squance proceeded to connect it with the quarter-inch pipe end in the closet.
On Friday, November 4,1932, the college had finished installing the necessary piping, and the gas was connected with Bunsen burners on the laboratory table. There were four burners on each side of the table. Ward tried out the burners and could only get about an inch of flame. The outlet of the Bunsen burner was not much larger than the lead in a pencil. He then tried out an old burner which had an opening about three-eighths of an inch in diameter and got a flame two or three feet high. Ward said he then thought there might be a leak. The gas was on for about an hour. He then turned it off at the cut-off in the closet, and Squance turned it off at the service cock in the cabinet outside of the building and gave the key to the cabinet to Ward. The gas was not turned on again until Wednesday, November 9, at about ten o’clock a. m.
Appellant telephoned the office of the company at Winslow, complained only of lack of pressure, and requested that an expert be sent. Several calls were *548placed, appéllant showing impatience at the delay in getting an expert to check the equipment, the company saying that it had been too busy to send a man. Finally, one Meharry was sent on Wednesday, November 9, at "ten o’clock a. m. He frankly confessed that he was not an expert, did not know anything about Flamo equipment, but had brought with him a new regulator and gauge and offered to, and did, disconnect the regulator then in use and substituted a new regulator and gauge. A test was then made of the Bunsen burners, and no improvement in the pressure was found.
Meharry then shut off the service cock in the outside cabinet, locked it, and handed Squance the key. Meharry told him he didn’t know what the matter was, but would send some expert who would be able to tell. Squance handed the key to Ward and said he -did not open the cabinet again. Ward said he did, trying out the Bunsen burners again on Wednesday afternoon for not to exceed fifteen minutes. He did it for the purpose of showing the students how weak the flame was. About 4:30 p. m. that day, an explosion occurred which wrecked one end of the building, seriously damaged the rest of it, and destroyed most of the equipment, apparatus and laboratory.
One witness for appellant, an architect, testified that the explosion came from beneath the basement floor. Squance testified that the Flamo man who installed it outside the building showed them how to operate it. He said to turn on the valve and then watch the indicator, it would run .up to six ounces and stop right there. The man also said “if you get any leaks whatever in the lines, the gauge will drop back.” Squance said he understood from that explanation that, when there was nothing open and the gauge dropped back *549from six ounces, it would indicate that there were leaks in the line. He further testified:
“We had the service cock turned on and tried out the burners, and couldn’t get a big enough flame, and when we couldn’t make the burners work satisfactorily, we tried the test he gave us for finding leaks. He said the gauge would drop back in case there was a leak. We tested for leaks by shutting off the burners, turning the service cock on, and the gauge stayed right up there, then we turned the service cock off and the gauge still stayed right up there. The gauge never showed any leak either way. It stood up both times. We did not turn off the valve on top of the cylinders. After we got through we turned it off at the service cock, so there would be no gas coming into the pipe line, and called the Standard Oil Company.”
In its complaint, appellant charges that respondent company was grossly negligent in the following particulars :
“(a) By virtue of the negligence of the defendant Allison, he at the time being the defendant company’s agent and representative;
“(b) In causing to be installed defective equipment with a defective pressure gauge which failed to indicate escaping gas;
“(c) In failing promptly to respond to the request made by the plaintiff of it that a skilled representative be immediately sent to the plaintiff’s school for the purpose of inspecting the equipment and locating and correcting the defects;
“(d) In finally sending an employee who was unskilled in the use of Flamo gas and equipment and who opened the valves of the tank and permitted gas to escape in large quantities while supposedly testing the equipment.”
At the trial, the judge correctly held that, where the evidence as to negligence amplified the allegations of negligence in the complaint, the complaint would be deemed amended to conform to the proof.
*550Appellant’s assignment of error is that the court erred in taking the case from the jury at the conclusion of appellant’s evidence and nonsuiting and dismissing the action. Appellant argues the matter under two heads: one, that there was negligence on the part of respondents from which a jury could find that such negligence was the proximate cause of the damages; and two, that appellant was not guilty of contributory negligence as a matter of law.
Appellant asserts that the only possible explanation of the explosion was that the gas in the narrow space between the floors, coming’ out into the furnace room through the holes in the concrete wall which separated them, thus coming in contact with oxygen, had been raised to .a kindling point by a spark from the motor which operated the sump pump.
To begin with, if appellant’s theory of the cause of the explosion is correct, as above stated, that means that Squance, its own maintenance man, was so careless in the installation of the piping underneath the basement floor as to leave some opening uncovered or some connection loose or unsealed. He, being the plumber, as he testified, who installed the pipes, knew, or in the exercise of reasonable care should have known, that the opening existed and that escaping gas was bound to occur. Of these facts, it is manifest that respondents had no notice or knowledge.
However, on the question of negligence on the part of respondents, the general rule, as stated in the annotations following the two decisions of Helm v. Manufacturers Light & Heat Co., 86 W. Va. 628, 104 S. E. 59, 25 A. L. R. 240, and McClure v. Hoopeston Gas & Electric Co., 303 Ill. 89, 135 N. E. 43, 25 A. L. R. 250, is:
*551‘ ‘ Generally speaking, a gas company which does not install the pipes in a customer’s building, and which has no control over them, is in no way responsible for the condition in which they are maintained, and consequently is not liable for injuries caused by a leak therein, of which it has no knowledge.”
See, also, note in 90 A. L. R. 1088, to the same effect.
Our court has said:
“Illuminating gas is a dangerous thing when it is not under control, and it is incumbent upon those who deal in it as an article of merchandise to use care commensurate with its harmful nature. The degree of care must be such as an ordinarily prudent person would exercise under like circumstances in managing such an article. [Citing cases.]” Senske v. Washington Gas & Electric Co., 165 Wash. 1, 4 P. (2d) 523.
In the instant case, it must be noted that respondent oil company is not a general distributor of illuminating gas through mains and pipes to customers. Apparently, there was no such service available to appellant. The rule .announced under the facts controlling the Senske case, supra, has very little bearing on this case.
Considering the case at bar, however, on a parity with the liability of a gas company producing and distributing its products through mains and pipes, such company would not be liable unless it had some duty of inspection of appliances on the premises of the consumer and some notice or knowledge which it had, or ought to have had, of the appliances and apparatus on the premises of the consumer. The cases cited by appellant, of which Consolidated Gas Co. v. Getty, 96 Md. 683, 54 Atl. 660, 94 Am. St. 603, United Oil Co. v. Roseberry, 30 Colo. 177, 69 Pac. 588, and Atkinson v. Wichita Gas Co., 136 Kan. 854, 18 P. (2d) 127, are typical, were all eases where the company *552supplying the gas had some such duty of preliminary inspection or responsibility to repair.
In this case, by the undisputed evidence, no apparatus furnished by the company was proven to be defective. The explosion did not occur outside of the building in the tanks or cabinet furnished by respondent company. No inspection was possible on the part of the representative of respondent of the pipes under the floors covered by cement. See Schmeer v. Gaslight Co., 147 N. Y. 529, 42 N. E. 202, 30 L. R. A. 653,
If the gas pipes and fittings are the property of the consumer and there is no contractual duty resting on the gas company to inspect, the consumer, by application for gas service, assumes the burden of inspecting and maintaining the pipes and fittings on his premises in a manner reasonably suited to meet the required service, and the company has the right to assume that these duties have been performed by the applicant. Okmulgee Gas Co. v. Kelly, 105 Okla. 189, 232 Pac. 428.
The company, having had no notice that the pipes were in defective condition on the consumer’s own premises, had the right to assume, in the absence of such notice, that the owner of the building had seen to it that the pipes were in proper repair and that the gas could not escape therefrom. Reid v. Westchester Lighting Co., 236 N. Y. 322, 140 N. E. 712, 29 A. L. R. 1247.
Other cases and texts announcing the same principle are: Smith v. Pawtucket Gas Co., 24 R. I. 292, 52 Atl. 1078, 96 Am. St. 713; Fleegar v. Consumers’ Power Co., 262 Mich. 537, 247 N. W. 741; Lewis v. Southern California Gas Co., 92 Cal. App. 670, 268 Pac. 930; Greed v. Manufacturers’ Light & Heat Co., 238 Pa. 248, 86 Atl. 95; Helm v. Manufacturers Light & Heat Co., supra; 12 R. C. L. 909, and 45 C. J. 954.
*553Under the above authorities, we fail to see where there was any actionable negligence, either gross or ordinary, on the part of respondents.
However, even if there was negligence on the part of respondents, we agree with the trial court that it was just as much negligence of Ward, acting for appellant, as it was of the agent of the oil company. We further consider that there must have been some previous negligence on the . part of Squance. Ward knew all of the conditions and knew the dangers of the element with which he was dealing. He was, apparently, of rather more than ordinary intellectual capacity. He was qualified and experienced in both science and chemistry. From the first installation, the Bunsen burners failed to function, which proved there must be a leak. If there were any dangers from leaks, he knew where they would be most apt to occur, which, manifestly, must have been under the concrete floor, where respondent and its representatives could not possibly have inspected and over which they had no control.
When the representative of respondent called on Wednesday noon, before the explosion occurred, before he left, he did the one thing that from that time on would protect the building’ from any damage by reason of escaping gas. He turned the valve off in the equipment which respondents furnished. After that agent had left, Ward turned the gas on again for a period not to exceed fifteen minutes between 1:15 and 1:30 p. m. The explosion occurred at about 4:30 p. m.
Ward, who was the principal representative of appellant, having mental capacity of a high order, must have utilized it to its fullest extent or appellant is not entitled to recover. 20 R. C. L. 107. See, also, 45 C. J. 947-953.
*554From tbe undisputed evidence, it must be concluded that tbe agents of appellant were guilty of sucb contributory negligence that appellant cannot recover. Tbe judgment is affirmed.
Mitchell, Beals, Blake, and Main, JJ., concur.