Raymond P. Fix, a resident of Rock Rapids, Iowa, was injured by a gas explosion which occurred November 9, 1950, in the basement of the house in which he was then living. He was at the time attempting to light the pilot burner of an automatic gas water heater manufactured by the American Radiator & Standard Sanitary Corporation, a Delaware corporation, and intended for use with liquefied petroleum gas, also called bottle gas.

Fix brought this action against the corporation upon the claim that it had manufactured the heater, and that the explosion and his injuries were due to its negligence in failing to equip the heater with a proper safety control and in failing to discover that the control attached to the heater was defective and the heater dangerous.

The defendant in its answer admitted that it had manufactured the water heater and that it was designed for use with liquefied petroleum gas. The defendant stated that it had not manufactured the safety control valve (referred to in the evidence as the safety valve or safety pilot), but had assembled it with other parts of the heater. The defendant admitted that the explosion had occurred and that the plaintiff had been injured, but denied that the defect in the control valve, if any existed, was due to any negligence on the part of the defendant.

The issues were tried to a jury. At the close of the evidence the defendant moved for a directed verdict upon substantially the following grounds: (1) that the plaintiff had failed to show that the safety valve or safety pilot, designed to shut off the gas from the heater if the pilot light went out, was defective at the time the heater left the defendant's factory; (2) that there was no evidence of any failure on the part of the defendant to properly inspect the device; (3) that the evidence conclusively showed that the safety pilot was operating properly after the heater was installed in the basement of the home occupied by the plaintiff; and (4) that the evidence was at least as consistent with the defendant's theory that the safety pilot had become defective after it left the defendant's factory as with the plaintiff's theory that the safety pilot was defective when it left the factory.

The court refused to direct a verdict for the defendant. The jury returned a verdict for the plaintiff. This appeal followed.

The defendant asserts that the District Court erred in denying the defendant's motion for a directed verdict, and erred with respect to certain rulings on evidence and with respect to the instructions to the jury.

Much of the evidence was undisputed. What the plaintiff endeavored to prove, and what he claims he did prove, was that at the time of the explosion the heater was in the same condition as when it left the factory; that the safety pilot or thermostatic control, the function of which was to shut off the gas from the heater automatically when the pilot flame went out, was out of adjustment and permitted gas to escape from the main burner and flow into the basement; and that the fact that the safety pilot was out of adjustment could and would have been discovered by the defendant had it used ordinary care in testing the safety pilot before it left the factory.

The evidence showed that the heater, after it was completed, was crated by the defendant at its factory. It was sold to a jobber in Sioux City, Iowa. The jobber sold the heater to Harry Rasmussen, doing business at Rock Rapids, Iowa, as Rasmussen Tin Shop. He received it on or about April 18, 1950. It was still uncrated. He sold it almost immediately to Earl Smith for installation in the house owned by him in which the plaintiff was living. The heater was taken to the house by employees of Rasmussen, uncrated there, and installed in the basement. They made the necessary water connections and ran a vent pipe from the heater to the chimney to carry off fumes from the burner.

The gas supply for the heater, constisting of two tanks or bottles of gas located at the north side of the house, was connected to the heater by Floyd Nyenhuis, who was employed by Dick DeGroot, the proprietor of the Modern Home Store in Rock Rapids, from whom the plaintiff bought the bottles or tanks of gas. There were two gas lines or pipes running from the gas supply into the house, one to the stove in the kitchen and the other to the water heater in the basement. After Nyenhuis had connected the heater with the gas supply, he lit the pilot flame according to the directions which accompanied the heater.

The heater was of the conventional type, cylindrical in shape, standing upright on legs, with the fire box in its base. It was provided with two burners: the main burner which heats the water, and a pilot burner adjacent thereto which supplies the flame which ignites the main burner if the thermostat which regulates the temperature of the water calls for heat. That occurs when the water in the water tank of the heater falls below a predetermined temperature.

The entire thermal unit attached to the heater, which regulates and controls the flow of gas to the burners, was manufactured by the Titan Valve Manufacturing Co. and furnished by it to the defendant. The defendant attached it to the heater. The unit has a hand valve with three positions indicated by a dial, namely 'off', 'pilot', and 'on'. When this valve is on the 'off' position, no gas flows to either of the burners in the fire box and the heater is inoperative. When the valve is on 'pilot', the gas flows to the pilot burner alone if the pilot tube is open. When the hand valve is in the 'on' position, the gas will flow to the main burner provided the pilot flame is burning or if the main burner tube is open all of the way to the burner. The unit has two thermostatic controls or thermal tubes. One is inserted in the water and is adjustable. It calls for heat and turns on the gas to the main burner when the water is below the predetermined temperature and the hand valve is in the 'on' position. This thermostat turns off the gas from the main burner when the water in the tank has reached the desired temperature.

At the lower end of the unit which controls the flow of gas to the burners is another thermostat or thermal tube which extends into the fire box and which, unless heated to a high temperature by the pilot flame, will, if properly adjusted, shut off all gas from both burners. In other words, this device is intended and should be adjusted to permit no gas to flow to either burner if the pilot flame is unlighted. Thus so long as the pilot flame is burning, the gas can flow to the main burner when heat is called for, but none can flow if the pilot flame is out unless the thermostat in the fire box is defective or out of adjustment. This thermostat and its component parts constitute the safety valve or safety pilot.

In order to start the heater operating when the safety pilot is properly adjusted, it is necessary first to turn the hand valve to the 'pilot' position. It is then necessary to press a push-button, one of the parts of the safety pilot, which operates a valve-set at the lower end of the gas control unit. This will open the gas tube of the pilot and permit the pilot burner to be lighted. When it is lighted, the push-button must be held in position long enough to allow the pilot flame to heat the thermal tube, which is just above the flame, to a high temperature. This may take as long as two or three minutes. When the thermal tube has reached the proper temperature, the push-button and the valve-set which it operates may be released and the flame will continue to burn. If the push-button-operated valve-set is released before the tube has been sufficiently heated, a spring will force the valve-set into its original position, and the pilot flame will go out. By the same token, if the pilot flame, for any reason, goes out while the heater is operating, and the thermal tube cools off, the safety pilot, if properly adjusted, will permit no gas to flow to either burner, even if the hand valve is in the 'on' position. After the pilot flame is burning and the thermal tube in the fire box is hot, the hand valve is placed and kept in the 'on' position and the thermostatic controls connected with the heater take over its operation and keep the water at a substantially uniform temperature so long as there is a supply of gas.

If the thermal tube of the safety pilot is defective or out of adjustment, so that it does not close off the gas from the burners when the pilot flame is out, then gas will flow from the main burner whenever the thermostat which regulates the heat of the water is calling for more heat and the hand valve is in the 'on' position.

The fact that Nyenhuis lit the pilot burner in the conventional way after he had connected the gas to the heater did not conclusively show that the safety pilot tube was or was not in proper adjustment, because, if the safety pilot permitted gas to flow into the pilot burner without the push-button being pressed, the burner would have lighted regardless of the position of the push-button.

Only one tank which supplied gas to the heater was used at a time. When that tank became empty, it was shut off manually by a valve at the top of the tank, and the full tank was then turned on, the empty tank being replaced by the Modern Home Store. The record shows that on three occasions, May 18, 1950, July 3, 1950, and August 15, 1950, after the heater was installed, the pilot flame went out, due to the exhaustion of a tank, and the plaintiff was required to relight the pilot burner. This was done immediately after he had switched to the full tank. On those three occasions he relighted the burner safely, following the usual procedure. On these occasions, unless the thermostat in the tank of the heater had been calling for heat, the lighting of the pilot burner in the usual way would not have disclosed a defect, if any existed, in the adjustment of the thermal tube of the safety pilot.

On October 13, 1950, the pilot flame went out again, due to one of the tanks of gas becoming empty. After connecting the full tank, the plaintiff went down to light the pilot burner. He placed the hand-valve on 'pilot' and pushed the button of the safety pilot, holding a match in the fire box. He did not know whether he held the match right on the pilot burner, because he did not get down on his knees to look into the fire box. He did not get a light, so he turned the hand-valve to the 'off' position and went for help. About ten or fifteen minutes later Gerrit Byker, of the Modern Home Store, came over and lit the burner. Byker told the plaintiff 'that maybe I (the plaintiff) had not held the button in long enough which was why I had failed to light the heater and that was my trouble.' After October 13, 1950, the plaintiff had no occasion to relight the burner until the time of the explosion.

On November 8, 1950, one of the tanks of gas became empty. The plaintiff's wife had switched over to the full tank, but had not attempted to relight the pilot burner of the heater. She told the plaintiff about that when he came home, but he did not go down to light the heater that night. There was no hot water the next morning. He went down to the basement, but smelled no gas. When he attempted to relight the pilot burner, the explosion occurred, indicating that the safety pilot had not shut off the gas to the main burner after the pilot flame went out on the previous day.

Byker, who serviced and installed furnaces for the Modern Home Store, testified, as a witness for the plaintiff, relative to his experience with the heater on October 13, 1950. Byker's testimony, on direct examination, was in part as follows:

'I went to the basement through a trap door in the kitchen and found a hot water heater similar to the one here in court. I lit the pilot light by opening that little door and lighting a match and putting it on 'pilot' and pushing that little button and hold it until the gas would flow to the pilot. The trouble I had in lighting it was that it seems to me that it took a long time to light it. I pressed the button in, held it in, and put a match where the pilot light was and got it lit. I made no adjustment on the control on the device at that time, and did not touch any part of the mechanism down here, which is partially in the fire box, and which you have identified as the thermal tube and pack nut. I did not touch that part of the mechanism at all. I wouldn't know how long I was down there. About the time to light it. I would say two or three minutes. It seems to me I met Mr. Fix as I came up, or about ready to come up and he was standing at the head of the stairs. I don't remember being in the basement on any other occasion before the explosion occurred. I made no adjustment on this occasion of the controls on the hot water heater.'

On cross-examination, Byker stated that the first thing he did was turn the manual control valve to the 'pilot' position, and that he did not recall what position that valve was in before he turned it; that he then struck a match, put it in the burner door, pushed the button in on the safety pilot, and held the match about where the pilot tube was. The further cross-examination, and redirect examination, or Byker follows:

'(Cross-Examination.)

'Q. And in that process did you use a wrench or pliers or screw driver or anything like that? A. The way I recall, it took a little time before it (the pilot light) would stay on and I likely did, I just tapped on the pipe. I don't know what I did.

'Q. 'That pipe' don't mean much for the record. Will you step down and show the jury where you tapped? A. Maybe like this, and then hit on here.

'Q. You tapped on this main pipe that goes down from the thermostat to the safety pilot? A. Yes, the safety valve.

'Q. Some of us have been calling it safety valve and some of us have been calling it safety pilot, but I mean the unit that goes into the main burner? A. That is right.

'Q. Now, then, why did you tap on the pipe? A. It seems to me it took a little long to light. Maybe I was too much in a hurry. I don't know.

'Q. You must have had some purpose in tapping on the pipe? A. Just one of them things.

'Q. Now, then, can you give me any estimate how long it took you to light it? A. No. I would say two or three minutes, maybe.

'Q. Does the fact that it took you quite a while to light it have any significance to you from your practical operation of these things. A. No. I would not say. I have had one that has taken as long as four minutes that I have timed it, to light.

'Q. You mean other occasions on other heaters? A. Other occasions, yes.'

'I can't recall what I told Mr. Fix when I met him on the stairs. I had not reached in the main burner tube with any wrench or pliers and had not made any change in the adjustment, and if someone did, don't know who it was. I would say it took me about the same time to light that unit that is usually the case in lighting those units. When I left there, the flame was operating properly as far as I could observe. The main burner light was on. When I leaned down to light it, I leaned far enough so that I could look in there and see the pilot tube.

'Q. When you pushed the button in did you keep holding it in or did you release it for a while and the light go out, or what happened? A. I held it in until the light stayed on. If you let up on it, it goes out.

'Q. You got a light right away? A. Yes.

'Q. But of course, you had to hold it until this unit became hot enough so it would stay on? A. That is right.

'Q. Just so I am perfectly clear about it, Mr. Bykers, you didn't release the button at any time, or did you? What I am getting at, did you release the button at any time and then let the light go out? A. I think I did once.

'Q. You think it did once. So then you had to hold it in a longer time, next time? A. Yes.

'Mr. Braithwaite: That is all.

'Redirect Examination.

'By Mr. Evans:

'Q. You just think so, but, of course, you don't remember distinctly as to just what happened? You just went down and lit it. Is that it? A. Yes.

'Mr. Briathwaite: Objected to as leading and suggestive.

'The Court: It is leading, but there is an answer in the record. Let it stand.

'Q. It was about the same as when you told us you didn't know just where the main control valve was, that is, whether it was on 'pilot' or 'off' or just where it was, you don't remember? A. I don't recall, no.

'Mr. Evans: That is all.

'(Recross Examination.)

'By Mr. Braithwaite:

'Q. When you said you thought the light went out once, you were giving us your best and honest recollection?

'Mr. Evans: Objected to for the reason the witness has already told us that is merely what he thought. This is repetition.

'The Court: Sustained.

'Mr. Braithwaite: That is all.

'Mr. Evans: That is all.'

The testimony of Byker that he had had to hold the push-button in two or three minutes to get the pilot burner to stay lighted, that he though he released the push-button once and the light went out, and that because the light would not stay on he had rapped, apparently with a wrench or pliers, on the pipe leading to the safety pilot, and the testimony of the plaintiff that Byker told him that his trouble was that he 'maybe' had not held the button in long enough, was strongly indicative that when Byker attempted to light the pilot flame of the heater on October 13, the safety valve or safety pilot was not out of adjustment. If it was a fact that it took Byker several minutes to get the pilot burner to stay lighted, and that he once released the push-button and the burner went out, then the safety pilot thermal tube was not out of adjustment at that time and therefore not out of adjustment when the heater left the factory of the defendant. If the thermal tube had been out of adjustment on October 13, there would have been no difficulty in lighting the pilot burner and no reason why it would not have continued to burn, regardless of the push-button.

It is apparent that plaintiff's counsel realized that Byker's testimony on cross-examination was damaging to the plaintiff's case, and that that was the reason why on redirect examination, by asking leading questions, he sought to discredit so much of that testimony as was favorable to the defendant by getting Byker to say that he did not recall what occurred. It is also apparent that that is what counsel for the plaintiff had in mind when he objected to defendant's counsel asking Byker, on recross-examination, whether the evidence he had given represented his best recollection.

The defendant's objection to the first question asked Byker on his redirect examination was well taken, but the question had already been answered. The defendant did not ask that the answer be stricken. To the second leading question, which was equally objectionable, no objection was taken. While we think that the court, after these leading questions had been asked and answered, should have permitted the defendant to cross-examine Byker as to whether his account of what occurred with respect to his relighting of the pilot light of the heater was his best recollection, the failure of the court in that regard would hardly rise to the dignity of reversible error; but compare, Alford v. United States, 282 U.S. 687, 692, 51 S. Ct. 218, 75 L. Ed. 624; Kroger Grocery & Baking Co. v. Stewart, 8 Cir., 164 F.2d 841, 843-845.

We think it was not error for the court to deny the defendant's motion for a directed verdict. It was conceded that, after the explosion, the thermal tube of the safety pilot was out of adjustment; that it would not shut off the gas from the main burner when the pilot flame was out. There was evidence tending to show that no one who had had access to the heater when and after it was installed had tampered with the adjustment of the thermal tube of the safety pilot and that nothing had occurred between the time the heater left the factory and the time it was examined after the explosion adequately to account for the tube being out of adjustment. It is true that the defendant's testimony showing that it and the manufacturer of the safety pilot had and maintained at all times a rigid system of inspection of all safety pilots manufactured and assembled, which inspection would have disclosed any failure of the safety pilot to operate properly, would have justified the jury in finding that there was no failure to inspect, but the witnesses who gave this testimony were not able to say definitely that this particular heater had been subjected to this inspection. Moreover, the evidence as to the inspection was not that of completely disinterested witnesses, and if the jurors believed that the thermal tube had not been tampered with since it left the factory and that its adjustment after the explosion was the same as when it left the factory, they were justified in finding that the safety pilot had not been properly inspected.

We think, however, that the District Court committed prejudicial error in failing to give the following instruction requested by the defendant:

'You are instructed that if you find from the evidence that at any time after the hot water heater involved in this action was installed in the basement of the home occupied by plaintiff that the safety valve was working properly, then it has been conclusively established that the plaintiff has failed to prove that the safety valve was inherently dangerous at the time it left the factory of the defendant, and plaintiff cannot recover in this action.'

At the close of the court's charge to the jury, the defendant objected to the failure of the court to give this requested instruction. The court then gave a further unobjectionable instruction which did not, either in words or substance, cover this request. The failure of the defendant to make further objection to the refusal of the court to give the requested instruction did not, we think, under the circumstances, preclude the defendant from asserting error in this regard. The language of the request was plain and the defendant was entitled to have it given. The requested instruction related to a vital issue in the case. The evidence showed that the thermal tube, after the explosion, was found to be in perfect condition, but slightly out of adjustment. If it was in proper adjustment when Byker lit the pilot burner in October, as the jury reasonably might have found, that, as we have previously stated, demonstrated conclusively that it was properly adjusted then and must have been in proper adjustment when it left the factory. There were marks on the thermal tube when it was removed and inspected after the explosion which the jury could have found to indicate that someone had tampered with the adjustment of the thermal tube after the heater left the defendant's possession.

The principles of law applicable to this case are, we think, the same as those announced in the leading case of MacPherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050, L.R.A. 1916F, 696, and which were applied by this court in Egan Chevrolet Co. v. Bruner, 8 Cir., 102 F.2d 373, 122 A.L.R. 987. See also O'Rourke v. Day & Night Water Heater Co., 31 Cal. App. 2d 364, 88 P.2d 191, and Restatement of the Law of Torts, Sec. 395, page 1073. It was the duty of the defendant to use due care in constructing the heater in suit and to see that it was equipped with a safety pilot which would completely shut off gas from the burners when the pilot flame was out, so that the heater could safely be used for its intended purpose and would not be a menace to those who used it or were near it. It was also the duty of the defendant to make all the ordinary and necessary tests to ascertain whether the safety pilot was in proper working condition and properly adjusted before the heater left the factory.

The jury in this case was required to determine from the evidence whether it was more probable that the safety pilot was out of adjustment when it left defendant's factory than that someone had tampered with the adjustment of the safety pilot thereafter. The evidence shows that if a safety pilot is improperly adjusted, so that gas can flow to the burners when the pilot flame is out, that fact is easily discoverable by simple tests. If a manufacturer fails to discover it, the fault is his and not that of the ultimate purchaser or user of the dangerous heater.

The case of O'Rourke v. Day & Night Water Heater Co., 31 Cal. App. 2d 364, 88 P.2d 191, upon which the defendant relies in contending that it was entitled to a directed verdict, is distinguishable. In that case there was apparently no evidence that the safety pilot had been defective at the time the heater was assembled or for some time thereafter and no evidentiary basis for any inference that the thermal tube of the safety pilot was in the same condition at the time of the explosion as when it left the factory, or that it had not been properly inspected by the manufacturer. In the instant case it is certain either that the safety pilot was improperly adjusted when it left the factory or that its adjustment had been tampered with thereafter. Under the evidence, we think it was for the jury, and not for the court, to say which of the two permissible but conflicting inferences should be drawn.

Other matters relating to the court's charge, of which the defendant complains, will not recur upon a new trial and need not be discussed.

The judgment appealed from is reversed, and the case remanded for a new trial.