HOLMES, Circuit Judge.

The appellee instituted this action against the appellant to recover property damages on a policy of insurance against an explosion. The complaint alleged in substance that a fire hydrant of the City of Atlanta, located on the sidewalk in front of the bank's premises, exploded; and that, as a direct result of said explosion, water rushed from said hydrant and flooded the basement floor of the bank with mud and water, damaging plaintiff's property. The appellant averred, first, that there was no explosion; second, that the proximate cause of the loss was mud and water; and, third, that the policy expressly excepted from coverage any explosion originating within steam boilers, pipes, fly-wheels, engines, and machinery connected therewith and operated thereby.

Appellant contends that, in order to have an explosion, you must have something that has changed its chemical or molecular structure under heat or chemical reaction. It argues that neither water nor cast-iron explodes; that a rifle, a firecracker, a steam boiler explodes, but that "we do not even dream" that a "bottle of drinking water explodes." The appellee argues that the primary question before us is whether there was substantial evidence before the jury to support its finding that the loss or damage was the result of an explosion within the meaning of the policy in suit, and that implicit in the verdict is the finding of the generally accepted definitions of the term explosion which bring this loss within the extended coverage of the policy.

In the trial upon the merits, the jury returned a verdict for the appellee; and there was evidence to warrant a finding of the following facts: A street cleaner opened a fire hydrant and, after flooding the street, started to close it. As he was turning the hydrant key in the closing direction, the hold-down device at the top of the hydrant leaped into the air and fell free of the hydrant. At that instant water ceased coming from the nozzle of the hydrant, and noises in the nature of a thud and like the sound of a firecracker, and other sounds, came from underground at the bottom of the hydrant, which series of sounds frightened the street cleaner. While no further water came from the top of the hydrant, it came in great quantities from the base thereof, a portion of which had been blown out; the water escaping from this rupture at the hydrant's base forced its way through the concrete floor of appellee's basement, damaging the articles located therein.

The underground rupture of the hydrant was caused by a water-hammer, which is an instantaneous build-up of pressure by the sudden stoppage of the flow of water. Calculations show that at the moment of this rupture, explosion, or water-hammer, the pressure in this hydrant jumped from an estimated 50 pounds per square inch to 615 pounds per square inch, which was more than was necessary to break and did break the casting forming the elbow at the base of the hydrant. This excessive pressure occurred very suddenly, probably "in a few thousandths of a second, following the closing of that valve." The bursting of this same casting in tests made above ground after the accident sounded like a rifle shot; it was a loud cracking sound. The excessive energy of the moving water had to be released in some way, and it was dissipated by the bursting of the casting. The increased and excessive pressure in the elbow at the base of the hydrant was due to the valve being slammed shut, which caused the alleged explosion.

An explosion is a rapid, sudden, and violent expansion of air or relinquishment of energy, causing a rupture and accompanied by a loud noise, not necessarily extremely loud. Fire is perhaps the principal cause of explosions, but they may be produced without the aid of fire. A flywheel may explode from excessive speed if its governor breaks. Even severe frost often causes trees to explode. It has been said that there is no difference in ordinary usage between explode and burst; the ordinary idea is that the explosion is the cause while the rupture is the effect. 1 Bouvier's Law Dictionary, Rawle's Third Revision, page 1161 citing United Life, Fire Marine Ins. Co. v. Foote, 22 Ohio St. 340, 348; Evans v. Columbian Ins. Co., 44 N.Y. 146, 150. The acts that constitute an explosion within the meaning of the extended coverage provision of this policy must be determined from the language of the policy, the ordinary uses of the word, the common experience of men, and their general notions in matters of this sort. One of the definitions in Webster's dictionary for the verb explode is to "burst noisily." In Louisville Underwriters v. Durland, 123 Ind. 544, 24 N.E. 221, 223, 7 L.R.A. 399, the court said that the bursting of a boiler and the explosion of a boiler were one and the same thing. In 35 C.J.S., Explosion, page 215, note 3, it is said that an explosion may be produced by the sudden liberation of air from an air gun.

In addition to the ordinary usage and commonly accepted meaning of this crucial word, we must look to the policy itself to ascertain what the parties meant by it, because their intention is the pole star in this litigation. The original form of this policy was intended for fire insurance, and contained a provision that, unless otherwise provided, the company should not be liable for loss occurring as a result of explosion or riot unless fire ensued, and in that event for loss by fire only. Attached to this policy, for an additional premium, was an endorsement that extended the coverage to include, among other things, loss or damage by explosion. This endorsement specifically provided that the insurer should not be liable for any loss by explosion originating within steam boilers, pipes, etc., connected therewith and operated thereby; and that any other explosion clause made a part of this policy was superseded by this endorsement. It is clear that the alleged explosion in the present suit did not originate within steam boilers or pipes connected therewith, and that any other explosion clause made a part of this policy was superseded by the extended coverage endorsement.

The appellant argues that the only similarity between a water-hammer and an explosion is that they both generally make a noise. There is another similarity, we think, which is that in both instances the bursting is caused by excessive pressure, and the pressure is caused by pent-up energy. The difference is that in steam boilers the excessive pressure is caused by heat or fire, while in water-hammers it is caused by energy generated by the speed of moving water. It is the sudden release of energy in both cases that causes the explosion. Energy is known to science only by its effects. The appellant having charged an additional premium for coverage to include explosion, with an exclusionary clause as to explosions originating within steam boilers, etc., and having provided that every other explosion clause in the policy was superseded by said endorsement, we have here a policy provision for insurance against loss or damage by explosion in the unrestricted use of the word, except only as to an explosion originating within steam boilers, pipes, etc., connected therewith and operated thereby.

Expert witnesses in this case testified that the cause of the rupture was the sudden build-up of pressure resulting from the slamming shut of the valve. The force of the water displaced a portion of the concrete floor in the basement, making a hole therein about four feet wide and five feet long. To take an illustration in miniature: Everyone knows the difference between a blow-out and a puncture in an automobile tire. The blow-out is a small explosion, while ordinarily the puncture merely causes a leak. The question in this case is whether there was a big leak or an explosion in the cast-iron elbow at the base of the hydrant. Cast-iron is rather brittle; it shatters like glass. Ordinarily it breaks with a large report; sometimes it sounds like a cannon shot. When this elbow broke there was no slow tearing apart; "very definitely it went all at once," said one expert witness. When asked if this casting failed because of old age, the witness replied: "Well, now, that is a strange question because as far as the cast iron is concerned, there is no old age unless it corrodes"; that he examined this piece and it did not have any particular amount of corrosion on it. This particular piece was also laboratorily tested and found to be well above standard quality; it took a load of 7500 pounds to break it; as to pressure, it was also well above standard quality.

Without trying to explain how the kinetic energy, when suddenly stopped, was changed into potential energy, and this in turn almost instantaneously converted into static pressure beyond the yielding point of the cast-iron elbow, we think that it was the sudden releasing of energy that caused this explosion. Whether the energy was generated by molecular changes, breaking down of the atom, chemical reaction, or the movement of water, is immaterial so far as this insurance coverage is concerned, provided only that it did not originate within the exclusionary clause of the extended coverage provision of the policy.

The question of proximate cause in this case was for the jury. We find no reversible error in the record, and the judgment appealed from should be affirmed. See L.L. Olds Seed Co. v. Commercial Union Assurance Co., 7 Cir., 179 F.2d 472; Lever Bros. Co. v. Atlas Assurance Co., 7 Cir., 131 F.2d 770; 22 Am. Jurisprudence, p. 126; 45 C.J.S., Insurance, § 885, page 950.

Affirmed.