Plaintiff brought this diversity action on a policy of insurance issued by defendant to plaintiff which insured the latter’s premises against damages resulting from various specified risks, including explosion. The parties submitted a “Stipulation of Uncontested Facts” and filed cross motions for summary judgment. The issue presented for decision was whether or not the damage to plaintiff’s premises was caused by an “explosion” under the terms of the policy. The district court ruled in favor of defendant and plaintiff appeals.
On January 7, 1965, plaintiff’s premises were in major part destroyed by the occurrence here in question, involving an autoclave used for the curing of concrete blocks. The autoclave was a cylindrical pressure vessel having an internal diameter of eight feet and a length of approximately ninety-two feet. The unit was designed to build up internal steam pressure with a maximum of 150 lbs. per square inch (psi). The safety valve was set at 142 lbs. psi. At 1:30 a. m. on the date in question the steam was turned on and the autoclave began building up pressure. At about 4:15 a. m., when the pressure was at 124 lbs. psi, a loud noise was heard. Inspection of the premises disclosed that the spherical door of the autoclave had become detached from the unit and had been propelled several hundred feet “by the suddenly released steam pressure,” greatly damaging the building and its contents. The autoclave itself was propelled approximately one hundred feet in the opposite direction “by the remaining force” of steam pressure, causing further destruction. The parties stipulated that plaintiff’s minimum damages are $88,-589.66 and that “[a]s far as is presently known, no sudden increase in internal pressure occurred.”
The district court, in finding for the defendant, relied on this latter statement in holding that no explosion occurred. The court, in a memorandum opinion, stated:
“We are dealing with the word ‘explosion’ in the way in which it is commonly accepted and understood. The common intelligent and logical understanding of an explosion of a container by sellers and buyers of industrial insurance1 is a sudden vio*1325lent bursting or breaking apart of the container caused by abnormal pressure. It is what happens when pressure suddenly is built up in a container to a point sufficiently in excess of the normal capacity of the container to contain it, so that the container suddenly and violently gives way. * * *
“Under this definition, the defendant must prevail. It is undisputed that the autoclave’was designed to operate under a steam pressure of 150 lbs. psi, and that at the time of the occurrence the pressure was 124 lbs. psi, or 26 lbs. psi under the rated capacity of the unit. Thus, the cause of the occurrence was not a sudden build up of pressure to a point in excess of the normal capacity of the container to contain it.”
We respectfully disagree with the district court’s conclusion that no explosion occurred. The policy did not define the term “explosion,” but contained the following relevant provisions:
“Provisions Applicable Only to Explosion: Loss by explosion shall include direct loss resulting from the explosion of accumulated gases or unconsumed fuel within the firebox (or the combustion chamber) of any fired vessel or within the flues or passages which conduct the gases of combustion therefrom. However, this Company shall not be liable for loss by explosion, rupture or bursting of:
(a) steam boilers, steam pipes, steam turbines, or steam engines ; or
(b) rotating parts of machinery caused by centrifugal force;
if owned by, leased by or actually operated under the control of the Insured.
“The following are not explosions within the intent or meaning of these provisions:
(a) Concussion unless caused by explosion,
(b) Electrical arcing,
(c) Water hammer,
(d) Rupture or bursting of water pipes.”
The defendant conceded that liability was not denied on the basis of any of the specific exclusions contained in the policy, but contends that no explosion occurred because no sudden increase in internal pressure took place before the steam was violently released. Thus, the issue before this Court is whether or not such violent and sudden release of steam from a container operating at below maximum pressure is an “explosion,” as that term is commonly understood. In determining this issue, we must follow the law of Illinois, the state of the forum. Erie Railroad Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188 (1938).
The rule of construction of insurance contracts was stated by the Illinois Supreme Court in Canadian Radium and Uranium Corp. v. Indemnity Insurance Co., 411 Ill. 325, 332, 104 N.E.2d 250, 254:
“The construction to be given insurance contracts, like other contracts, should be a natural and reasonable one. * * * If the language is clear, the terms are to be taken and understood according to their plain, ordinary and popular sense. * * * But because the insurer is the one who prepares the contract, ambiguous or equivocal expressions whereby the insurer seeks to limit its liability will be construed most strongly against the insurer.”
The Court went on to look at the dictionary definition of the word there in *1326question, “accident,” as well as the Illinois cases dealing with the term.
In the instant case, we are directed by the parties to the dictionary as well as various cases defining “explosion.” Webster’s New International Dictionary, Unabridged (2d ed. 1954), defines explosion as “a violent bursting or expansion with noise, following the sudden production of great pressure, as in the case of explosives, or a sudden release of pressure, as in the destruction of a steam boiler.” The Random House Dictionary of the English Language, Unabridged (1967), defines explosion as “a violent expansion or bursting with noise, as of gunpowder or a boiler (opposed to implosion).” Webster’s Third New International Dictionary (1966) added the following definition to the term: “a violent expansion or bursting that is accompanied by noise and is caused by a sudden release of energy from a very rapid chemical reaction, from a nuclear reaction or from an escape of gases or vapor under pressure (as in a steam boiler).”2
Thus, as defined in the dictionary, an explosion includes an occurrence where a great and expanding pressure is suddenly released. It is the violent expansion itself which constitutes the explosion, and none of the above authorities limit the occurrence of such expansion to the inside of a container. Thus, the ignition of an uncontained combustible material, such as a plastic explosive or gunpowder, would cause an explosion since there would be a sudden and great increase in pressure, an expansion, in the area surrounding the ignition. Likewise, the internal pressure of a boiler, if suddenly released by a defective enclosure, would cause an explosion — “an escape of * * * vapor under pressure” (Webster’s Third Edition, supra) —whether or not such pressure was built up slowly or rapidly. Again, it is the sudden creation or release of the great amount of pressure which causes the violent expansion which is the explosion.
Defendant relies heavily upon two cases to support its contention and the district court’s conclusion that without a “sudden increase” in pressure within the autoclave, no explosion occurred. The first of these cases is Hulcher Soya Products v. Millers’ Mutual Fire Insurance Association, 5 Ill.App.2d 235, 124 N.E.2d 570 (abstract opinion, Third District, 1955). The Appellate Court there dealt with the issue of whether or not damage to a grain elevator was caused by an “explosion,” one of the perils insured against by the defendants. The evidence showed that a wall of the elevator had been forced out “accompanied by a loud noise, dust, grain, broken and crushed concrete, and some violence as shown by the pieces of concrete being thrown away from the building approximately 200 feet.” The court held that the verdict for the plaintiff — which found that the damage was caused by an explosion — was against the manifest weight of the evidence since “the record disclose [d] no direct evidence of the sudden development of an internal force or that the incident characterized by plaintiff as an explosion was accompanied by a sudden or rapid expansion of air;”3 and that circumstantial evidence could not establish the fact of explosion since the existence of an inconsistent fact, or the complete non-existence of the first fact, could be inferred from the evidence with equal certainty. In reciting the facts of the case, the appellate court listed *1327the stipulated jury instructions, which included as the definition of “explosion” the criteria advanced by defendant in the instant controversy:
“1. A sudden accident [al] violent bursting or breaking;
2. That such bursting or breaking, if any, was caused by a suddenly developed internal force;
3. That such bursting or breaking, if any, was accompanied by a sudden or rapid expansion of air; and,
4. That such bursting or breaking was accompanied by a sharp noise or report.”
The district court in the instant case discussed Hulcher and concluded that it was not relevant since the instruction was there stipulated by the parties and not in issue, stating:
“The obvious reason that the explosion instruction itself was not litigated in Hulcher Soya is that if you in fact have a grain explosion, it will, of course, be of the type where the bursting is ‘caused by a suddenly developed internal force.’ In a grain dust explosion, minute particles of grain are suspended in air, and, upon ignition, result in a combustion explosion. Had this been a silage explosion, we might have had proper litigation of the issue, since in a silo explosion bacteria cause the silage to ferment, releasing carbon dioxide, raising internal pressure, and ‘exploding’ the silo.
“In short, Hulcher concerns whether there was any evidence to support a given determination as to what constitutes an explosion rather than what that determination should have been in the first instance. The latter is the problem confronting me. Hulcher leaves me free to make a determination of first impression in this jurisdiction.”
We agree that the fact that the Hulcher court cited an uncontested jury instruction does not make the definition contained in that instruction binding on us as the law of Illinois. However, even were we to accept defendant’s contention that the Appellate Court “inherently accepted the definition of ‘explosion’ contained in the jury instruction,” we are of the opinion that the facts of Hulcher distinguish it from the case at bar. The collapse of the grain elevator in Hulcher involved no combustion, gaseous expansion, or expansion of any other substance, as does the instant case, but rather involved the mere, albeit violent, spilling of a solid mass produced by the force of gravity against the wall of the elevator. It was the plaintiff’s failure to prove such an expansion in Hulcher which caused the Appellate Court to reverse.
Moreover, in so holding, the court stated that there was “no direct evidence of a sudden development of an internal force or that the incident characterized by plaintiff as an explosion was accompanied by a rapid or sudden expansion of air.” (Emphasis supplied.) Thus, the Illinois Court recognized that “a sudden development of an internal force” was not the exclusive criteria of an explosion, as defendant contends; additionally, “a sudden rapid expansion of air” was contemplated by the court as another definition of the term. The Hulcher case, then, seems to support plaintiff’s contention in the instant case that a sudden and violent expansion, internal or external, may cause or become an explosion upon sudden release.
Defendant also cites Honeymead Products Company v. Aetna Casualty and Surety Company, 275 Minn. 182, 146 N.W.2d 522 (1966), to support its position that a sudden increase in internal pressure is requisite to an explosion. In that case, the Minnesota Supreme Court found that the rupture of a steel tank containing a large quantity of soy bean oil was not an “explosion,” but merely a rupture or bursting caused by a fracture in the tank. As with Hulcher, we feel the facts distinguish Honeymead from the instant case in that the rupture, while releasing a violent *1328flood of liquid, resulted in no expansion of either the air surrounding the liquid or of the liquid itself. It is the violent expansion of the steam which was suddenly released in the instant case which constituted the explosion. The very nature of steam which is compressed to far beyond its normal pressure distinguishes it from grain or oil in ■ its natural state. While the force of gravity may cause the latter substances to spill with violent force when support of the container is suddenly removed, it is the internal pressure of compressed steam which causes it, when suddenly released, to violently expand — to explode.
Defendant’s argument would seem to exclude the explosion of a steam boiler unless there was a sudden increase in internal pressure. The difference between the violent bursting of a steam boiler which is suddenly heated and one which gradually builds up an abnormal amount of pressure is beyond our comprehension. Had the autoclave in the instant ease slowly built up pressure in excess of its maximum 150 lbs. psi, resulting in a violent rupture, could defendant plausibly argue that no explosion occurred since the excessive pressure had not developed suddenly? The policy itself, by its inclusion of “explosion, rupture, or bursting of: (a) steam boilers, steam pipes, steam turbines, or steam engines * * * ” seems to contemplate the opposite result.4
Thus, in circumstances involving a container, an explosion occurs when the pressure inside the container exceeds the strength of the container and results in a sudden release of the pressure. This may happen either when a higher than normal internal pressure exceeds the natural breaking point of the container, or when there is a decrease in the natural breaking point of the container operating under normal internal pressure.5 Both are equivalent conditions resulting in an explosion which in turn results in the acceleration of the fragments or parts of the container.
In Lever Brothers Co. v. Atlas Assurance Company, 131 F.2d 770, 775 (7th Cir. 1942), this court quoted the lower court’s instruction that “Explosion means a sudden violent bursting or breaking caused by an internal force and accompanied by a sudden or rapid expansion of air and a sharp noise or report.” Later in the opinion, the Court quoted Webster’s definition as “substantially the same as the [lower] court’s * * * ‘a violent bursting or expansion, with noise, following the sudden production of great pressure * * * or a sudden release of pressure.” (Emphasis supplied.) In none of the cases cited by defendant do we find a requirement that in all cases of an explosion there must be a sudden increase in internal pressure. Rather, if one criterion can be found, it is “a sudden expansion of a substance.” See Hartford Fire Insurance Company v. Empire Coal Mining Company, 30 F.2d 794, 797 (8th Cir. 1929); Bolich v. Provident Life and Accident Insurance Company, 205 N.C. 43, 169 S.E. 826, 828; Peterson v. Royal Insurance Company, 251 N.C. 61, 110 S.E.2d 441, 443.
There is a recent case, curiously not cited by either party to the case at bar, Aetna Casualty and Surety Company v. Osborne-McMillan Elevator Company, 35 Wis.2d 517, 151 N.W.2d 113 (1967), which specifically rejected the insurer’s contention that “a sudden and violent increase in the internal active force” is required to constitute an explosion.
*1329“The increase of the internal force must manifest itself in a sudden and violent action such as a sudden and violent release by breaking the container, but to do this the increase of a nonexplodable substance like grain need not be sudden although in explodable substances there usually is a sudden buildup of pressure or force due to a chemical or nuclear reaction. We expressly stated [in an earlier opinion involving this controversy] the pressure required for an explosion to take place ‘need not be unusual, abnormal or suddenly produced.’ 26 Wis.2d 292, at 304, 132 N.W.2d 510. There is a distinction between a sudden and violent increase and a sudden and violent release. It is the sudden and violent release of the force which is an essential element of an explosion.” 35 Wis.2d at 524, 151 N.W.2d at 116.
In discussing the Honeymead case, relied upon by the insurer in that ease as in the instant case, the Court said:
“[T]he Minnesota supreme court held that the normal pressure from the weight of the stored oil caused the brittle plates of the tank to fracture and the fracture was not due to an explosion ‘ * * * caused by a sudden or violent increase of pressure from within the tank.’ The court also said there was no evidence ‘ * * of any sudden or violent increase of internal pressure commonly associated with an explosion.' 275 Minn, at 191, 192, 146 N.W.2d at 528, 529. From this language the insurance companies argue the court found an essential element of an explosion to be a sudden or violent increase in internal pressure. We think not. The language used must be taken in reference to the fact that there was no increase in the pressure of the soybean oil, sudden or otherwise, over that of its weight and position in the tank. The decision quoted with approval from our first opinion wherein we pointed out an element of explosion was the sudden and violent release of the pressure from the container. 146 N.W.2d at 528. We do not read the Honeymead decision as holding an explosion can only occur if there is a sudden or violent increase of internal pressure in the tank.” 35 Wis.2d at 525, 151 N.W.2d at 116-117.
Thus, it was the sudden and violent expansion of a substance — steam, upon its sudden release — which was the explosion in the instant case, and the absence of such sudden expansion of any substance which caused the courts to rule that no explosions took place in Hulcher and Honeymead.
We disagree with defendant’s additional contention that there is no foundation in the stipulation of facts for plaintiff’s statement that “the occurrence was caused by a violent expansion of steam confined under pressure of 124 lbs. psi, which was suddenly released from an autoclave. * * * ” The stipulation contains the statement that
“the entire door [of the autoclave] was propelled forward by the suddenly released steam pressure. The south door being blown off, released the pressure force at that end while considerable internal pressure still remained at the north end of the cylinder. This remaining force propelled the entire unit northward a distance of approximately 100 feet.”
Finally, defendant contends that the rule of strict construction of insurance contracts against the insurer is inapplicable to the instant case since it claims no ambiguities exist in the language of the policy. We note that the policy did not define the term “explosion” as used therein and that vigorous contention exists over the definition of that term.6 The words of the II-*1330linois Supreme Court in the Canadian Radium case, supra, 411 Ill. at 334-355, 104 N.E.2d at 256, are particularly appropriate:
“In any event, it is apparent that, at the very least, ambiguity exists as to the content and scope of the word ‘accident,’ as employed in the policy issued to plaintiff. Under the well-settled rule, the meaning of the word most favorable to the insured should be accepted, since the insurer prepares the contract. (Mosby v. Mutual Life Ins. Co., 405 Ill. 599, 92 N.E.2d 103, 18 A.L.R.2d 1054; Roth v. Kaplowsky, 393 Ill. 484; Pioneer Life Ins. Co. v. Alliance Life Ins. Co., 374 Ill. 576, 30 N.E.2d 66.) This rule is particularly applicable where, as here, the very name of the plaintiff should have informed the insurer of the likelihood of occurrences such as the one described in Mary Moore’s complaint against plaintiff in the Federal Court, and induced it, if it did not intend to insure against such a risk, to exclude specifically from coverage any liability because of injury, sickness or disease caused by radio-activity.”
Thus, in the instant case, the nature of plaintiff’s business, the production of concrete blocks, and the conspicuous presence on plaintiff’s premises of four huge autoclaves put the insurer on notice that plaintiff probably intended to insure itself against an occurrence such as that which took place on January 7, 1965. The defendant, being the drafter of the insurance contract, could have excluded such coverage; but, having failed to do so, and having left the pertinent term of the contract undefined, the court should adopt the meaning of that term most favorable to the •insured.
For the foregoing reasons, the judgment of the district court is reversed and the case remanded with directions to enter summary judgment for the plaintiff on the issue of liability.
Reversed and remanded.
. We note that the district court’s defining explosion in terms of “intelligent and logical understanding of an explosion of a container by sellers and buyers of indus*1325trial insurance,” seems to be a departure from the “ordinary and popular” standard applied in 'most jurisdictions, including Illinois. See Mitchell v. Potomac Insurance Co., 183 U.S. 42, 52, 22 S.Ct. 22, 46 L.Ed. 74 (1901); Canadian Radium and Uranium Corp. v. Indemnity Ins. Co., 411 Ill. 325, 104 N.E.2d 250 (1954).
. Among the many cases which have referred to Webster’s definition of explosion are: Collins v. City of Chicago, 321 Ill. App. 73, 91, 52 N.E.2d 473 (1943) (dictum); Lever Bros. Co. v. Atlas Assurance Co., 131 F.2d 770, 775 (7th Cir. 1942); Hartford Fire Insurance Company v. Empire Coal Mining Company, 30 F.2d 794, 797 (8th Cir. 1929); Chicago, R. I. & P. R. Co. v. Aetna Insurance Co., 180 Kan. 730, 308 P.2d 119, 124-125 (1957).
. The evidence disclosed that no explosive gas and no excess pressure could have ' developed in the elevator.
. The rule of expressio unius est exclusio alterius might well apply here in that by not excluding the rupturing or exploding of a steam autoclave, such an event may be deemed to be included.
. A tragic example of such an explosion were the incidents involving the British Comet aircraft, which exploded in flight, under normal cabin-operating pressure, due to metal fatigue in the fuselage. See D. Anderson, How the Comet Mystery Was Solved, 62 Aviation Week 26-39 (February 14, 1955).
. Indeed, “the cases admit that ‘explosion’ is a term insusceptible of fixed definition.” American Cas. Company of Reading, Pa. v. Myrick, 304 F.2d 179, 183 (5th Cir. 1962). Cf. Hartford Fire Insurance Company v. Empire Coal Mining Company, 30 F.2d 794, 797 (8th Cir. 1929).