This is an appeal from a judgment awarding plaintiff damages for injuries received in an explosion which *703occurred in a service station owned by 0. M. McCormick and which was situated at Greenfield on Highway 99 in Kern County. The jury returned a verdict in favor of plaintiff in the sum of $6,500. This was reduced by the trial judge to $4,500 on motion for new trial.
Defendant thus states the questions involved in this appeal upon which it relies for a reversal of the judgment. “1. Did respondent produce any evidence that any natural gas leaking from appellant’s gas service pipe caused the explosion resulting in injuries to respondent ? Answer: No. 2. Does not the evidence of both parties prove that this explosion was caused by gasoline vapors? Answer: Yes.” This presents the sole question of the sufficiency of the evidence to support the verdict and judgment. Incidentally, there is argued the question of error in refusing to strike out evidence of a witness who had smelled gas in the service station near the place of the explosion, on the ground that it related to a time too remote from the time of the explosion.
In our statement of facts we will set forth those which tend to support the verdict and judgment.
The service station was situated at the southwest corner of Union Avenue (Highway 99) and Taft Avenue. It faced on the highway. Defendant’s gas main extended north and south in Highway 99, and its service pipe, about sixty feet long, led into the station and paralleled its south wall about 43 inches from it at a depth of about thirty inches. The ground was permeable loam. The service pipe was installed in 1931. The service station had been remodeled, the work having been completed about three weeks prior to the explosion which occurred at about 7:15 o’clock on the evening of June 8, 1940.
There were eleven underground storage tanks for gasoline and diesel oil in the station grounds south of the station. During the remodeling the vent pipes from these tanks were brought to the south wall of the building in newly dug trenches which crossed over the gas service pipe. The vents were carried up the side of the building and were enclosed in a false flue which was made of a wooden frame covered with plaster. There were openings from this flue so that the gasoline and gas fumes could escape into the partitions and walls of the service station. Two of the vent pipes did not extend to the top of the flue. One ended about eighteen inches and the other between eight or ten feet below the top. The shorter *704vent pipe connected with a storage tank into which three hundred gallons of gasoline had been dumped about fifteen minutes before the explosion.
During the remodeling the area south of the service station had been covered with cement which reached to the south wall. The high point of this cement slab was where it joined the false flue, the cement sloping both south and southeast from that point. The bottom of the flue rested on dirt. The explosion occurred when plaintiff turned a switch to light the service station. The switch was in a room adjoining the flue. The explosion blew out the walls of this and other rooms and blew plaintiff through a wall.
Plaintiff offered evidence that the smell of gas had been noticeable near the south wall of the service station for several years prior to the explosion; that complaint of this escaping gas had been made to a meter reader and collector of the predecessor of defendant; that the last complaint was made about two years before the explosion.
McCormick testified as follows:
“Q. Directing your attention, Mr. McCormick, to this area south of this room we have here, what has been described as the west wall of the old kitchen running from here a little bit east of what is now the present rooms A and B, and then the north room as I draw the pencil on Plaintiff’s Exhibit 1, and south towards the most easterly wall of the old kitchen and then at this end of the room which is marked W-12 as drawn by one of the witnesses, the location of your old stove in that kitchen—I will ask you Mr. McCormick if immediately preceding this accident and for some time prior thereto, if, in the location of the outside wall of that old stove, that is immediately outside of this south wall, if you had noticed gas fumes there? A. Yes, sir. Q. And I will ask you if you ever reported those fumes to any of the gas company employees? A. I reported it to the meter reader and the gas collector. Q. Did anyone from the Gas Company ever come out and make an investigation ? A. I saw one fellow testing it, and I believe they were out there three or four times, but I never saw but the one fellow testing it, and I do not know who he was. Q. And what did they do when they made their investigation ? A. They checked the fittings around the meter and checked the meter, and of course the meter would not show anything. We had no leaks that could be found. It had been that way for two years and was getting worse. Q. On *705how many occasions would you say you reported it to the Gas Company? A. I talked several times with Mr. Walford and with the meter reader, and we kept checking to find out ourselves, but we could not find it.” (Emphasis added.)
Plaintiff testified as follows: “Q. When you have indicated that you smelled natural gas, the odor of natural gas, on the previous occasions you have indicated, it was in the region of the old meter box, is that correct ? A. Yes. Q. And was it around that area, around there, about a considerable distance or quite a distance, or what? A. Well, most anywhere in the garage part back there, you could smell it. Q. You are referring to the old garage ? A. Yes sir. Q. And at that time then within that area where you smelled that gas there was no cement slab at that time—that was prior to the reconstruction? A. No sir, there was no cement. Q. And after the cement slab was placed, your testimony is you didn’t smell any natural gas? A. That is right.”
On the day following the explosion defendant caused the gas service pipe to be uncovered and tested for leaks. One test was made by applying soap suds to the surface of the pipe. Any escaping gas would cause bubbles to arise in the soap suds. McCormick testified that he saw bubbles on the portion of the pipe immediately south of the room in which the explosion occurred. This evidence was corroborated by other testimony.
Defendant’s witnesses testified to several tests of this pipe. They admitted finding small leaks from a point forty-three feet east of the west end of the pipe to its connection with the main in Highway 99. They found no leaks in the west forty-three feet of the pipe.
Defendant also caused the ground around the service pipe and the walls of the service station to be tested with a combustible gas indicator which is an instrument that causes a red dial to become exposed when gas vapors in combustible quantities are discovered. A witness for plaintiff testified that the red dial was exposed when the test was made of the ground over the service pipe near the place of the explosion. According to defendant no combustible gases were found either at that point or in the walls or partitions of the service station. Apparently the instrument could not distinguish between gasoline vapors or gas in combustible quantities.
*706Experts for both parties agree that gasoline vapors are heavier than air and descend when released; that gas is lighter than air and rises when it escapes confinement; that both vapors will spread when the opportunity is presented. They agree that both gasoline fumes and gas when mixed with proper proportions of air are highly explosive. Here the agreement of the experts ends, and their testimony as to the probable cause of the explosion, whether from gas escaping from the service pipe, or gasoline vapor escaping from the short vent pipe, is in sharp and irreconcilable conflict.
It was the opinion of plaintiff’s expert that the explosion was in all probability caused from gas. It was his theory that gas escaped from the leaks in the service pipe, rose into the soft and permeable earth, followed the newly dug trenches (containing the vent pipes) under the concrete slab to its highest point at the false flue and escaped into the bottom of that flue in the soft dirt around the vents, rose in the flue and escaped through the openings and cracks in it into the walls and partitions of the service station where it was ignited by a spark from the switch on one of these walls when contact was made by plaintiff attempting to turn on the lights.
This witness had several reasons for his belief that the explosion was caused by the ignition of the escaping gas rather than from the ignition of gasoline fumes. The one of his several reasons which is most easily understood by a lay mind is this: That the ceilings of the small room containing the switch, and the room adjoining it, were lower than the ceilings of the other rooms of the service station, thus forming an enclosed well; that the ceiling of this well was covered with oil smudge and lint, both of which were easily ignitible; that there was much exposed wood on the inside of this well; that gasoline fumes explode with a flash which will ignite easily combustible material and char wood; that if gasoline vapors had escaped from the short vent pipe during the process of filling the gasoline tank those vapors would have filled this well and would have exploded; that the oil smudge and lint were not ignited and there was no charring of the exposed wood in this well. He drew the conclusion that the great probabilities were against the theory of the explosion having been caused by gasoline fumes and in favor of its having been caused by gas escaping from the service pipe.
Defendant’s experts were more positive that the explosion could not have been caused by escaping gas but was caused by gasoline vapors being forced from the storage tank during *707the filling operation and out of the end of the short vent pipe whence they settled in the flue and spread into the walls and partitions of the service station to be ignited by the spark from the electric switch.
Thus we have two conflicting theories of the cause of the explosion, with the jury accepting one as true and rejecting the other. While we might conclude from a study of the cold record that the theory advanced by the witnesses for defendant seems more probably correct and supported by better reason we cannot disturb the verdict and judgment unless we can reach the conclusion that the evidence supporting the judgment is inherently improbable or unworthy of belief. This we cannot do.
The correct rule that must guide us is set forth in Juchert v. California Water Service Co., 16 Cal. (2d) 500 [106 P. (2d) 886], as follows:
“The rule is not, as is sometimes contended and occasionally stated by the courts, that in order for an inference to be established by circumstantial evidence it must be the only inference which can fairly or reasonably be drawn from the circumstances relied on, or that if other inferences may reasonably be drawn from the facts in evidence, the evidence does not support the inference sought to be deduced from it.
“As stated in Robertson v. Weingart, 91 Cal. App. 715, 723 [267 Pac. 741], said principle is more applicable as a rule to guide a jury in its deliberations upon the facts, than it is a rule to guide the appellate court in passing upon the sufficiency of the evidence. (Peters v. McKay, 136 Cal. 73 [68 Pac. 478].) . . .
“A recent pronouncement of this court in the case of Hamilton v. Pacific Elec. Ry. Co., 12 Cal. (2d) 598, 600, 602 [86 P. (2d) 829], states the true position of an appellate court in cases where several inferences may be drawn from the evidence presented to a jury or trial court:
“ ‘It may be confidently declared that, founded upon the evidence, the jury not only is authorized to make any logical and reasonable deduction, but also that the jury is the exclusive judge of the weight and value of the inference that may be drawn by it; furthermore, as corollary thereto, that any attempt on the part of an appellate court to draw an inference of fact constitutes “a usurpation of the province of the trial court”. The fact that some inference other than that which has been drawn by a jury may appear to an ap*708pellate tribunal to be the more reasonable, affords no sufficient reason for disturbing the inference in question. ’ ” (See, also, Robertson v. Weingart, 91 Cal. App. 715 [267 Pac. 741]; Bellon v. Silver Gate Theatres, Inc., 4 Cal. (2d) 1 [47 P. (2d) 462]; Raggio v. Mallory, 10 Cal. (2d) 723 [76 P. (2d) 660].)
Defendant attacks the evidence of McCormick as so contradictory and uncertain as to be unreliable. There are many uncertainties and some contradictions in his testimony. Under the circumstances here those matters went to his credibility as a witness and the weight to be given his evidence. Those questions may be argued to the jury but cannot prevail here after that body evidently accepted as true portions of his evidence supporting the verdict and disregarded the uncertainties and contradictions in it. The latter went principally to the times he smelled the odor of gas, whether before or after a date in 1938, or both before and after that date. While the witness may have been confused under severe cross-examination and certainly contradicted himself, we believe that the jury could have construed all his evidence as indicating that he smelled the gas both before and after the specified date.
The question of the ruling of the trial court on the motion to strike the evidence of McCormick on his smelling the odor of gas, is confused in the record which shows the motion was first made in chambers and that the trial judge said: “I will grant the motion and strike this out.” It does not appear that this ruling was ever made in open court nor was it communicated to the jury. Thereafter McCormick testified that he had noticed the smell of gas after 1938.
Defendant had portions of McCormick’s evidence transcribed and introduced this transcription in evidence. At the close of the case it moved to strike this evidence out. This motion was denied. We do not regard this ruling as prejudicial. (Brunig v. Pacific Gas & Electric Co., 140 Cal. App. 254 [35 P. (2d) 226].)
The judgment is affirmed.
Barnard, P. J., and Griffin, J., concurred.
A petition for a rehearing was denied July 7, 1942, and the following opinion was thereupon rendered: